jsahlen-imdb 0.1.0

data/README.markdown

@@ -0,0 +1,21 @@
+IMDb Parser
+===========
+
+Built this as a learning excercise, and because the existing libraries didn't work exactly as I wanted them to.
+
+This is an extremely early version, which will likely be severely refactored as it progresses.
+
+Search example
+--------------
+
+This is the only functionality that works right now.
+
+    require 'imdb'
+
+    results = IMDB::Search.new("howl")
+    results.each do |movie|
+      puts "#{movie.title} (#{movie.year}) [#{movie.id}]"
+      unless movie.aka.empty?
+        movie.aka.each { |aka| puts "  a.k.a. #{aka}" }
+      end
+    end

data/lib/imdb.rb

@@ -0,0 +1,19 @@
+module IMDB
+  require 'open-uri'
+  require 'rubygems'
+  require 'hpricot'
+  require 'cgi'
+  require 'iconv'
+  require 'json'
+
+  FILES=%w{search movie}
+  FILES.each { |f| require File.join(File.dirname(__FILE__), 'imdb', f) }
+
+  TITLES_SEARCH_URL="http://www.imdb.com/find?s=tt&q="
+  TITLE_URL="http://www.imdb.com/title/"
+
+  def self.str_to_utf8(str)
+    Iconv.conv('UTF-8', 'LATIN1', str)
+  end
+
+end

data/lib/imdb/movie.rb

@@ -0,0 +1,83 @@
+module IMDB
+
+  class Movie
+    attr_accessor :id, :title, :year, :aka, :url, :director, :extra
+
+    def initialize(attributes={})
+      self.extra = {}
+
+      attributes.each_pair do |key, value|
+        self.instance_variable_set "@#{key}", value
+      end
+    end
+
+    def self.new_from_doc(doc)
+      movie = self.new
+      movie.parse_doc(doc)
+      movie
+    end
+
+    def self.new_from_id(id)
+      movie = self.new
+      movie.id = id.to_s
+      movie.get_full_details
+      movie
+    end
+
+    def get_full_details
+      doc = Hpricot(open(IMDB::TITLE_URL+CGI.escape(self.id)))
+      self.parse_doc(doc)
+      self
+    end
+
+    def parse_doc(doc)
+      akalink = (doc/"a[@href$='releaseinfo#akas']")[0]
+
+      self.id    = (doc/"a[@href$='fullcredits]")[0].attributes["href"][/title\/(.*?)\//, 1]
+      self.title = IMDB.str_to_utf8(CGI.unescapeHTML((doc/"h1").inner_html[/^(.*?)(?: <span)/, 1]))
+      self.year  = (doc/"h1 a[@href^='/Sections/Years/']").inner_html
+      self.aka   = akalink ? akalink.parent.inner_html.scan(/(?:>)([^<]*?)\(/).collect { |x| IMDB.str_to_utf8(CGI.unescapeHTML(x[0].strip)) } : []
+
+      parse_full_details(doc)
+
+      self
+    end
+
+    def parse_full_details(doc)
+      director_links = (doc/"#director-info/a")
+      writer_links   = (doc/"a[@onclick*='writerlist']")
+      tagline_header = (doc/"h5[text()='Tagline:']")[0]
+      plot_link      = (doc/"a[@href$='/plotsummary]")[0]
+      mpaa_link      = (doc/"a[@href='/mpaa']")[0]
+
+      if director_links.length > 0
+        self.director = director_links.collect { |l| IMDB.str_to_utf8(CGI.unescapeHTML(l.inner_html)) }
+      end
+
+      self.extra["writers"]     = writer_links.collect { |w| IMDB.str_to_utf8(CGI.unescapeHTML(w.inner_html)) } if writer_links
+      self.extra["tagline"]     = IMDB.str_to_utf8(CGI.unescapeHTML(tagline_header.parent.inner_html[/\/h5>(.+?)(<|$)/m, 1].strip)) if tagline_header
+      self.extra["plot"]        = IMDB.str_to_utf8(CGI.unescapeHTML(plot_link.parent.inner_html[/\/h5>(.+?)<a/m, 1].strip)) if plot_link
+      self.extra["mpaa_rating"] = IMDB.str_to_utf8(CGI.unescapeHTML(mpaa_link.parent.parent.inner_html[/\/h5>(.+)$/m, 1].strip)) if mpaa_link
+      self.extra["cast"]        = ((doc/"table.cast")[0]/"tr").collect { |tr| { "actor" => IMDB.str_to_utf8(CGI.unescapeHTML((tr/"td.nm")[0].inner_text)), "character" => IMDB.str_to_utf8(CGI.unescapeHTML((tr/"td.char")[0].inner_text)) } }
+
+      self
+    end
+
+    def id=(id)
+      @id = id
+      self.url = IMDB::TITLE_URL + id
+    end
+
+    def to_json(*a)
+      {
+        "id"       => self.id,
+        "title"    => self.title,
+        "year"     => self.year,
+        "aka"      => self.aka,
+        "director" => self.director,
+        "extra"    => self.extra
+      }.to_json(*a)
+    end
+  end
+
+end

data/lib/imdb/search.rb

@@ -0,0 +1,50 @@
+module IMDB
+
+  class Search
+    include Enumerable
+
+    attr_accessor :results
+
+    def initialize(title, opts={})
+      @limit = opts[:limit] || 0
+
+      if title =~ /^\s*$/
+        self.results = []
+        return self.results
+      end
+
+      doc = Hpricot(open(IMDB::TITLES_SEARCH_URL+CGI.escape(title)))
+
+      # Single match
+      unless (doc/"div#tn15.maindetails").empty?
+        self.results = [Movie.new_from_doc(doc)]
+
+      # Search result
+      else
+        links = (doc/"td > a[@href^='/title/']").delete_if { |a| a.inner_html =~ /^</ }
+        self.results = links.collect! do |a|
+          td = a.parent
+          movie = Movie.new
+
+          movie.id    = a.attributes["href"][/^\/title\/([^\/]+)/, 1]
+          movie.title = IMDB.str_to_utf8(CGI.unescapeHTML(a.inner_html))
+          movie.year  = td.inner_html[/<\/a>\s\((\d+)\)/, 1]
+          movie.aka   = td.inner_html.scan(/aka\s+<em>"([^"]+)"<\/em>/).collect { |x| IMDB.str_to_utf8(CGI.unescapeHTML(x[0].strip)) }
+
+          movie
+        end
+      end
+
+      self.results = self.results.slice(0, @limit) if @limit != 0
+    end
+
+    def each
+      self.results.each { |x| yield x }
+    end
+
+    def to_json(*a)
+      self.results.to_json(*a)
+    end
+  end
+
+end

metadata

@@ -0,0 +1,76 @@
+--- !ruby/object:Gem::Specification 
+name: jsahlen-imdb
+version: !ruby/object:Gem::Version 
+  version: 0.1.0
+platform: ruby
+authors: 
+- "Johan Sahl\xC3\xA9n"
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2009-08-14 00:00:00 -07:00
+default_executable: 
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  name: hpricot
+  type: :runtime
+  version_requirement: 
+  version_requirements: !ruby/object:Gem::Requirement 
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        version: "0"
+    version: 
+- !ruby/object:Gem::Dependency 
+  name: json
+  type: :runtime
+  version_requirement: 
+  version_requirements: !ruby/object:Gem::Requirement 
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        version: "0"
+    version: 
+description: A simple IMDb scraper
+email: johan.sahlen@gmail.com
+executables: []
+
+extensions: []
+
+extra_rdoc_files: 
+- README.markdown
+files: 
+- README.markdown
+- lib/imdb.rb
+- lib/imdb/movie.rb
+- lib/imdb/search.rb
+has_rdoc: false
+homepage: http://github.com/jsahlen/imdb
+licenses: 
+post_install_message: 
+rdoc_options: 
+- --charset=UTF-8
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.3.5
+signing_key: 
+specification_version: 3
+summary: A simple IMDb scraper
+test_files: []
+