imdb_parser 0.6.6

data/lib/imdb_parser/season.rb

@@ -0,0 +1,45 @@
+module Imdb
+  
+  class Season
+    attr_accessor :id, :url, :season_number, :episodes
+    
+    
+    def initialize(url)
+      @url = url
+      @season_number = url[-1,1]
+      @episodes = []
+    end
+    
+    # Returns numbers of episode
+    def episode_numbers
+      document.search("div div[@class*=season-filter-all filter-season-#{season_number}'] div[@class*=filter-all]").size rescue nil
+    end
+    
+    def episode(number = 1)
+      Imdb::Episode.new(number, document.search("div div[@class*=season-filter-all filter-season-#{season_number}'] div[@class*=filter-all]")[number - 1], self)
+    end
+      
+    def episodes
+      if @episodes.empty?
+        document.search("div div[@class*=season-filter-all filter-season-#{season_number}'] div[@class*=filter-all]").each_with_index do |node_episode, i|
+          @episodes << Imdb::Episode.new(i + 1, node_episode, self)
+        end
+      end
+      @episodes
+    end
+    
+    private
+    
+    # Returns a new Hpricot document for parsing.
+    def document
+      @document ||= Hpricot(Imdb::Season.find_by_season(@url))
+    end
+    
+    # Use HTTParty to fetch the raw HTML for this season
+    def self.find_by_season(url)
+      open(url)
+    end
+    
+  end # Season
+  
+end # Imdb

data/lib/imdb_parser/serie.rb

@@ -0,0 +1,24 @@
+module Imdb
+  
+  class Serie < ImdbBase
+    
+    # s = Imdb::Serie.new("0773262")
+    # e = s.seasons.first.episodes.first
+    def number_seasons
+      document.search("a[@href*=episodes#season']").size
+    end
+
+    def season_urls
+      document.search("a[@href*=episodes#season']").map { |link| url.gsub("combined","") + "episodes#season-" + link.innerHTML.strip.imdb_unescape_html } rescue []
+    end
+    
+    def seasons
+      s = []
+      season_urls.each_with_index do |url|
+        s << Imdb::Season.new(url)
+      end
+      s
+    end
+      
+  end  #Serie
+end # Imdb

data/lib/imdb_parser/string_extensions.rb

@@ -0,0 +1,28 @@
+require 'cgi'
+ 
+module Imdb #:nordoc:
+  module StringExtensions
+  
+    # Unescape HTML
+    def imdb_unescape_html
+      if String.method_defined?(:encode)
+        CGI::unescapeHTML(self.encode("UTF-8", 'ISO-8859-1'))
+      else
+        require 'iconv'
+        Iconv.conv("UTF-8", 'ISO-8859-1', CGI::unescapeHTML(self)) 
+      end
+    end
+  
+    # Strip tags
+    def imdb_strip_tags
+      gsub(/<\/?[^>]*>/, "")
+    end
+    
+    # Strips out whitespace then tests if the string is empty.
+    def blank?
+      strip.empty?
+    end unless method_defined?(:blank?)
+  end
+end
+
+String.send :include, Imdb::StringExtensions

data/lib/imdb_parser/top_250.rb

@@ -0,0 +1,10 @@
+module Imdb
+
+  class Top250 < MovieList
+    private
+    def document
+      @document ||= Hpricot(open("http://akas.imdb.com/chart/top"))
+    end
+  end # Top250
+
+end # Imdb

data/lib/imdb_parser/version.rb

@@ -0,0 +1,3 @@
+module Imdb
+  VERSION = '1.1.0'
+end

data/lib/imdb_parser.rb

@@ -0,0 +1,17 @@
+$:.unshift(File.dirname(__FILE__)) unless
+  $:.include?(File.dirname(__FILE__)) || $:.include?(File.expand_path(File.dirname(__FILE__)))
+
+require 'open-uri'
+require 'rubygems'
+require 'hpricot'
+
+require 'imdb_parser/imdb_base'
+require 'imdb_parser/episode'
+require 'imdb_parser/season'
+require 'imdb_parser/movie'
+require 'imdb_parser/serie'
+require 'imdb_parser/movie_list'
+require 'imdb_parser/search'
+require 'imdb_parser/top_250'
+require 'imdb_parser/string_extensions'
+require 'imdb_parser/version'

data/script/console

@@ -0,0 +1,11 @@
+#!/usr/bin/env ruby
+# File: script/console
+irb = RUBY_PLATFORM =~ /(:?mswin|mingw)/ ? 'irb.bat' : 'irb'
+
+
+libs =  " -r irb/completion"
+# Perhaps use a console_lib to store any extra methods I may want available in the cosole
+# libs << " -r #{File.dirname(__FILE__) + '/../lib/console_lib/console_logger.rb'}"
+libs <<  " -r /Users/matthieu/Dev/rails/perso/imdb/lib/imdb.rb"
+puts "Loading imdb gem"
+exec "#{irb} #{libs} --simple-prompt"

data/script/destroy

@@ -0,0 +1,14 @@
+#!/usr/bin/env ruby
+APP_ROOT = File.expand_path(File.join(File.dirname(__FILE__), '..'))
+
+begin
+  require 'rubigen'
+rescue LoadError
+  require 'rubygems'
+  require 'rubigen'
+end
+require 'rubigen/scripts/destroy'
+
+ARGV.shift if ['--help', '-h'].include?(ARGV[0])
+RubiGen::Base.use_component_sources! [:rubygems, :newgem, :newgem_theme, :test_unit]
+RubiGen::Scripts::Destroy.new.run(ARGV)

data/script/generate

@@ -0,0 +1,14 @@
+#!/usr/bin/env ruby
+APP_ROOT = File.expand_path(File.join(File.dirname(__FILE__), '..'))
+
+begin
+  require 'rubigen'
+rescue LoadError
+  require 'rubygems'
+  require 'rubigen'
+end
+require 'rubigen/scripts/generate'
+
+ARGV.shift if ['--help', '-h'].include?(ARGV[0])
+RubiGen::Base.use_component_sources! [:rubygems, :newgem, :newgem_theme, :test_unit]
+RubiGen::Scripts::Generate.new.run(ARGV)

data/spec/fixtures/search_kannethirey_thondrinal

@@ -0,0 +1,14 @@
+HTTP/1.1 302 Found
+Date: Wed, 14 Sep 2011 19:36:19 GMT
+Server: Server
+Cache-Control: private
+Location: http://akas.imdb.com/title/tt0330508/?fr=c2M9MXxsbT01MDB8ZmI9dXx0dD0xfG14PTIwfGh0bWw9MXxjaD0xfGNvPTF8cG49MHxmdD0xfGt3PTF8cXM9S2FubmV0aGlyZXkgVGhvbmRyaW5hbHxzaXRlPWFrYXxxPUthbm5ldGhpcmV5IFRob25kcmluYWx8bm09MQ__;fc=1;ft=1
+Cneonction: close
+Content-Type: text/plain
+Set-Cookie: uu=BCYv20qm93TIGyXIk82OGarTTY0NVT5VU-kcAuTWuF7jisHqWPSX_CZTPYQ7WQjzVL7ZKWixB0BzsuwCJzcB_MMipd8O954-acpbS8Y3T1M5aXGlGeyywn-2Nuref6-l0K6dSsteRd82ob6x23AoRiwaimQTTxgN_noULvCYOT10nBgmc4vMUx0q2k4P6f0uQLlk_B0ZmA8cgARvhEzG89yJmSkNc3DyBsvYuEhbR55mZDpJkhN8Xu4YdsZ6UpyCfe6G;expires=Thu, 30 Dec 2037 00:00:00 GMT;path=/;domain=.imdb.com
+Set-Cookie: session-id=349-6028979-0265356;path=/;domain=.imdb.com;expires=Mon, 12 Sep 2016 12:36:19 GMT
+Set-Cookie: session-id-time=1473708979;path=/;domain=.imdb.com;expires=Mon, 12 Sep 2016 12:36:19 GMT
+Vary: Accept-Encoding,User-Agent
+P3P: policyref="http://i.imdb.com/images/p3p.xml",CP="CAO DSP LAW CUR ADM IVAo IVDo CONo OTPo OUR DELi PUBi OTRi BUS PHY ONL UNI PUR FIN COM NAV INT DEM CNT STA HEA PRE LOC GOV OTC "
+Content-Length: 0
+

data/spec/fixtures/search_killed_wife

@@ -0,0 +1,14 @@
+HTTP/1.1 302 Found
+Date: Wed, 14 Sep 2011 19:36:12 GMT
+Server: Server
+Cache-Control: private
+Location: http://akas.imdb.com/title/tt0166222/?fr=c2M9MXxsbT01MDB8ZmI9dXx0dD0xfG14PTIwfGh0bWw9MXxjaD0xfGNvPTF8cG49MHxmdD0xfGt3PTF8cXM9SSBraWxsZWQgbXkgbGVzYmlhbiB3aWZlfHNpdGU9YWthfHE9SSBraWxsZWQgbXkgbGVzYmlhbiB3aWZlfG5tPTE_;fc=1;ft=7
+Cneonction: close
+Content-Type: text/plain
+Set-Cookie: uu=BCYp98AY1s7aXCCUXayHXaP1anZAIOWkruf4_bRbNI1LmvCib0NkZTf_Dyv1hcYK6kqssUGYU3sL-6vcaGm5YT9z7dWrODyvzjlW3yVi57Lohq9q2ycNs0wSY_a-1pHHPtCE3tyW-8cKs_WLAUOhogR8GnJWoKPz0rHycTek9j6SoyKO2qI0wIrg7EdjaE6cdRlqF9qOH9KEeqcucoVivBS1_BsRHbyzYKreavQ46kGSSn-aRQ1L6b2bPfK5giWlE-eV;expires=Thu, 30 Dec 2037 00:00:00 GMT;path=/;domain=.imdb.com
+Set-Cookie: session-id=651-6028972-5384590;path=/;domain=.imdb.com;expires=Mon, 12 Sep 2016 12:36:12 GMT
+Set-Cookie: session-id-time=1473708972;path=/;domain=.imdb.com;expires=Mon, 12 Sep 2016 12:36:12 GMT
+Vary: Accept-Encoding,User-Agent
+P3P: policyref="http://i.imdb.com/images/p3p.xml",CP="CAO DSP LAW CUR ADM IVAo IVDo CONo OTPo OUR DELi PUBi OTRi BUS PHY ONL UNI PUR FIN COM NAV INT DEM CNT STA HEA PRE LOC GOV OTC "
+Content-Length: 0
+