odeon_uk 1.1.5 → 2.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 5f884c0b0fef03d462a08ad8b2aa1faba780fcaf
-  data.tar.gz: 719bc6d321d5da808066734ec32d8c69556d619d
+  metadata.gz: 59c9821b04dd9e8262c25557b82d95b51a9fdf7d
+  data.tar.gz: b1950e1710b9a4bcda1dbe85611cb26ad3604c48
 SHA512:
-  metadata.gz: 309f7d5b07ad1ebb13c8186f4e36cf4d64c4755b2cc56873fb20a622ddd44dc47f5e52505be05aed3e054117d09f794e23454e7f3d87a65b069d9e858c38acbd
-  data.tar.gz: a0317f4d337308e42594380966133233732b15a4246596c7341696678d663547d771fff5afc6e7e9c399d08f5c59adff6d9f48ed8fe6c51115f9607feda96c0a
+  metadata.gz: dc7ef841b0748d993a87b43e77cebf818cefa48698a3cd461bb57178f96dc677b5f0014d488d49857e76512912034a5e7c37092dcf1109ee5eb5fddb6d5a81aa
+  data.tar.gz: 6be57bd2533423a9895ec6f9a4f6aed25c45721886ac244cbed0ab13720f3a31899d045f3d5fd09b936d556d732ece8f5636948b5787565fcb025e28506bc839

data/.travis.yml

@@ -1,4 +1,4 @@
 language: ruby
 rvm:
-  - 1.9.3
   - 2.0.0
+  - 2.1.2

data/CHANGELOG.md

@@ -1,4 +1,11 @@
-
+## 2.0.0, _22nd September 2014_
+
+- add website utility class
+- break up film parsing into internal classes
+- Film#at builds films not cinema
+- fixture update script
+- title sanitizer class
+- updated fixtures
 - Added changelog
 
 ## 1.1.3, _8th Nov 2013_

data/README.md

@@ -5,6 +5,7 @@ A simple gem to parse the [Odeon UK website](http://odeon.co.uk) and spit out us
 [![Gem Version](https://badge.fury.io/rb/odeon_uk.png)](http://badge.fury.io/rb/odeon_uk)
 [![Code Climate](https://codeclimate.com/github/andycroll/odeon_uk.png)](https://codeclimate.com/github/andycroll/odeon_uk)
 [![Build Status](https://travis-ci.org/andycroll/odeon_uk.png?branch=master)](https://travis-ci.org/andycroll/odeon_uk)
+[![Inline docs](http://inch-ci.org/github/andycroll/odeon_uk.png)](http://inch-ci.org/github/andycroll/odeon_uk)
 
 ## Installation
 

data/Rakefile

@@ -1,20 +1,24 @@
 #!/usr/bin/env rake
-require "bundler/gem_tasks"
+require 'bundler/gem_tasks'
 
 require 'rake/testtask'
 
 Rake::TestTask.new do |t|
   t.libs << 'lib/odeon_uk'
-  t.test_files = FileList['test/lib/odeon_uk/*_test.rb', 'test/lib/odeon_uk/internal/*_test.rb']
+  t.test_files = FileList[
+    'test/lib/odeon_uk/*_test.rb',
+    'test/lib/odeon_uk/internal/*_test.rb'
+  ]
   t.verbose = true
 end
 
-task :build do
-  system "gem build odeon_uk.gemspec"
+# http://erniemiller.org/2014/02/05/7-lines-every-gems-rakefile-should-have/
+task :console do
+  require 'irb'
+  require 'irb/completion'
+  require 'odeon_uk'
+  ARGV.clear
+  IRB.start
 end
 
-task :release => :build do
-  system "gem push odeon_uk-#{OdeonUk::VERSION}"
-end
-
-task :default => :test
+task default: :test

data/lib/odeon_uk/cinema.rb

@@ -1,8 +1,6 @@
 module OdeonUk
-
   # The object representing a cinema on the Odeon UK website
   class Cinema
-
     # @return [String] the brand of the cinema
     attr_reader :brand
     # @return [Integer] the numeric id of the cinema on the Odeon website
@@ -21,16 +19,13 @@ module OdeonUk
     def initialize(id, name, url)
       @brand = 'Odeon'
       @id    = id.to_i
-      @name  = name.gsub('London - ','')
-      @slug  = @name.downcase.gsub(/[^0-9a-z ]/,'').gsub(/\s+/, '-')
+      @name  = name.gsub('London - ', '').gsub(' - ', ': ')
+      @slug  = @name.downcase.gsub(/[^0-9a-z ]/, '').gsub(/\s+/, '-')
       @url   = (url[0] == '/') ? "http://www.odeon.co.uk#{url}" : url
     end
 
     # Return basic cinema information for all cinemas
     # @return [Array<OdeonUk::Cinema>]
-    # @example
-    #   OdeonUk::Cinema.all
-    #   #=> [<OdeonUk::Cinema brand="Odeon" name="Odeon Tunbridge Wells" slug="odeon-tunbridge-wells" id=23 url="...">, #=> <OdeonUk::Cinema brand="Odeon" name="Odeon Brighton" slug="odeon-brighton" chain_id="71" url="...">, ...]
     def self.all
       cinema_links.map do |link|
         new_from_link link
@@ -38,11 +33,12 @@ module OdeonUk
     end
 
     # Find a single cinema
-    # @param [Integer, String] id the cinema id of the format 71/'71' as used on the odeon.co.uk website
+    # @param [Integer, String] id the cinema id of the format 71/'71' as used on
+    # the odeon.co.uk website
     # @return [OdeonUk::Cinema, nil]
     # @example
     #   OdeonUk::Cinema.find('71')
-    #   #=> <OdeonUk::Cinema brand="Odeon" name="Brighton" slug="brighton" id=71 url="...">
+    #   #=> <OdeonUk::Cinema brand="Odeon" name="Brighton" slug="brighton" ...>
     def self.find(id)
       id = id.to_i
       return nil unless id > 0
@@ -55,7 +51,10 @@ module OdeonUk
     # @example
     #   cinema = OdeonUk::Cinema.find('71')
     #   cinema.adr
-    #   #=> { street_address: 'Kingswest', locality: 'Brighton', postal_code: 'BN1 2RE', country_name: 'United Kingdom' }
+    #   #=> { street_address: 'Kingswest',
+    #         locality: 'Brighton',
+    #         postal_code: 'BN1 2RE',
+    #         country_name: 'United Kingdom' }
     def adr
       {
         street_address: street_address,
@@ -68,15 +67,8 @@ module OdeonUk
 
     # Films with showings scheduled at this cinema
     # @return [Array<OdeonUk::Film>]
-    # @example
-    #   cinema = OdeonUk::Cinema.find('71')
-    #   cinema.films
-    #   #=> [<OdeonUk::Film name="Iron Man 3">, <OdeonUk::Film name="Star Trek Into Darkness">]
     def films
-      film_nodes.map do |node|
-        parser = OdeonUk::Internal::FilmWithScreeningsParser.new node.to_s
-        OdeonUk::Film.new parser.film_name
-      end.uniq
+      Film.at(id)
     end
 
     # The name of the cinema including the brand
@@ -99,7 +91,6 @@ module OdeonUk
       address_node.text.match(/\w+(\s\w+){0,}\s+(\w+(\s\w+){0,})/)[2]
     end
 
-
     # Post code of the cinema
     # @return [String]
     # @example
@@ -117,14 +108,7 @@ module OdeonUk
     #   cinema.screenings
     #   #=> [<OdeonUk::Screening film_name="Iron Man 3" cinema_name="Brighton" when="..." variant="...">, <OdeonUk::Screening ...>]
     def screenings
-      film_nodes.map do |node|
-        parser = OdeonUk::Internal::FilmWithScreeningsParser.new node.to_s
-        parser.showings.map do |screening_type, times_urls|
-          times_urls.map do |array|
-            OdeonUk::Screening.new parser.film_name, self.name, array[0], array[1], screening_type
-          end
-        end
-      end.flatten
+      Screening.at(id)
     end
 
     # Screenings for particular film
@@ -155,7 +139,7 @@ module OdeonUk
     private
 
     def self.cinema_links
-      links = parsed_sitemap.css('.sitemap > .span12:nth-child(4) a[href*=cinemas]')
+      links = parsed_sitemap.css('.sitemap a[href*=cinemas]')
       links.select { |link| link.get_attribute('href').match(/\/\d+\/$/) }
     end
 
@@ -171,7 +155,7 @@ module OdeonUk
     end
 
     def self.sitemap_response
-      @sitemap_response ||= HTTParty.get('http://www.odeon.co.uk/sitemap/')
+      @sitemap_response ||= OdeonUk::Internal::Website.new.sitemap
     end
 
     def address_node
@@ -179,23 +163,11 @@ module OdeonUk
     end
 
     def cinema_response
-      @cinema_response ||= HTTParty.get(@url)
+      @cinema_response ||= OdeonUk::Internal::Website.new.cinema(@id)
     end
 
     def parsed_cinema
       Nokogiri::HTML(cinema_response)
     end
-
-    def film_nodes
-      parsed_showtimes.css('.film-detail')
-    end
-
-    def parsed_showtimes
-      Nokogiri::HTML(showtimes_response)
-    end
-
-    def showtimes_response
-      @showtimes_response ||= HTTParty.get("http://www.odeon.co.uk/showtimes/week/#{@id}?siteId=#{@id}")
-    end
   end
 end

data/lib/odeon_uk/film.rb

@@ -1,5 +1,4 @@
 module OdeonUk
-
   # The object representing a film on the Odeon UK website
   class Film
     include Comparable
@@ -16,11 +15,20 @@ module OdeonUk
       @slug = name.downcase.gsub(/[^0-9a-z ]/,'').gsub(/\s+/, '-')
     end
 
+    # All currently listed films showing at a cinema
+    # @param [Integer] cinema_id id of the cinema on the website
+    # @return [Array<OdeonUk::Film>]
+    def self.at(cinema_id)
+      showtimes_page(cinema_id).to_a.map do |html|
+        new(screenings_parser(html).film_name)
+      end
+    end
+
     # Allows sort on objects
     # @param [OdeonUk::Film] other another film object
     # @return [Integer] -1, 0 or 1
-    def <=> other
-      self.slug <=> other.slug
+    def <=>(other)
+      slug <=> other.slug
     end
 
     # Check an object is the same as another object.
@@ -28,7 +36,7 @@ module OdeonUk
     # @return [Boolean] True if both objects are the same exact object, or if
     #   they are of the same type and share an equal slug
     # @note Guided by http://woss.name/2011/01/20/equality-comparison-and-ordering-in-ruby/
-    def eql? other
+    def eql?(other)
       self.class == other.class && self == other
     end
 
@@ -41,7 +49,17 @@ module OdeonUk
     # @return [Integer] hash of slug
     # @note Guided by http://woss.name/2011/01/20/equality-comparison-and-ordering-in-ruby/
     def hash
-      self.slug.hash
+      slug.hash
+    end
+
+    private
+
+    def self.screenings_parser(html)
+      OdeonUk::Internal::FilmWithScreeningsParser.new(html)
+    end
+
+    def self.showtimes_page(cinema_id)
+      OdeonUk::Internal::ShowtimesPage.new(cinema_id)
     end
   end
 end

data/lib/odeon_uk/internal/film_with_screenings_parser.rb

@@ -1,67 +1,154 @@
 module OdeonUk
-
   # Internal utility classes: Do not use
   # @api private
   module Internal
-
     # Parses a chunk of HTML to derive movie showing data
     class FilmWithScreeningsParser
+      # CSS selector for a film title inside an individual film HTML
+      NAME_CSS      = '.presentation-info h4 a'
+      # CSS selector for a group of showtimes inside an individual film HTML
+      SHOWTIMES_CSS = '.times-all.accordion-group'
 
-      # @param [String] film_html a chunk of html
-      def initialize(film_html)
-        @nokogiri_html = Nokogiri::HTML(film_html)
+      # @param [String] html a chunk of html
+      def initialize(html)
+        @html = html
       end
 
       # The film name
       # @return [String]
       def film_name
-        name = @nokogiri_html.css('.presentation-info h4 a').children.first.to_s
-
-        # screening types
-        name = name.gsub /\s+[23][dD]/, '' #remove 2d or 3d from title
-        name = name.gsub 'Autism Friendly Screening -', '' # remove autism friendly
-
-        # special screenings
-        name = name.gsub 'ROH -', 'Royal Opera House:' # fill out Royal Opera House
-        name = name.gsub 'Met Opera -', 'Met Opera:' # fill out Met Opera
-        name = name.gsub /Cinemagic \d{1,4} \-/, '' # remove cinemagic stuff
-        name = name.gsub 'UKJFF -', '' # remove UK Jewish festival prefix
-        name = name.gsub 'NT Live -', 'National Theatre:' # National theatre
-        name = name.gsub 'National Theatre Live -', 'National Theatre:'
-        name = name.gsub /\(Encore.+\)/, '' # National theatre, remove encore
-        name = name.gsub 'Bolshoi -', 'Bolshoi:' # bolshoi ballet
-        name = name.gsub /\([Ll]ive\)/, '' # remove bolshoi-style live
-        name = name.gsub 'Globe On Screen:', 'Globe:' # globe
-        if name.match /RSC Live/
-          name = 'Royal Shakespeare Company: ' + name.gsub(/\- RSC Live \d{1,4}/, '')
+        @film_name ||= TitleSanitizer.new(raw_film_name).sanitized
+      end
+
+      # array containing hashes of screening attributes
+      # @return [Array<Hash>]
+      def to_a
+        doc.css(SHOWTIMES_CSS).map do |node|
+          dimension_parser(node).to_a.map do |hash|
+            hash.merge(film_name: film_name)
+          end
+        end.flatten
+      end
+
+      private
+
+      def doc
+        @doc ||= Nokogiri::HTML(@html)
+      end
+
+      def raw_film_name
+        @raw_film_name ||= doc.css(NAME_CSS).children.first.to_s
+      end
+
+      def dimension_parser(node)
+        DimensionNodeParser.new(node)
+      end
+    end
+
+    # parses chunk of screenings for a particular screening dimension
+    class DimensionNodeParser
+      # CSS selector for a single screening inside a 'group of showtimes' HTML
+      SCREENING_CSS = '.show'
+      # CSS selector for showing technology for a 'group of showtimes' HTML
+      TECH_CSS = '.tech a'
+
+      # @param [Nokigiri::Node] node a Nokogiri node object
+      def initialize(node)
+        @node = node
+      end
+
+      # array containing hashes of screening attributes for dimension
+      # @return [Array<Hash>]
+      def to_a
+        screening_hashes.map do |hash|
+          hash.merge(
+            dimension: dimension,
+            variant: add_imax(hash[:variant])
+          )
         end
+      end
+
+      private
+
+      def add_imax(original)
+        imax? ? "#{original} imax".strip : original
+      end
+
+      def dimension
+        tech.match(/[23]d/)[0]
+      end
+
+      def imax?
+        !!tech.match(/imax/)
+      end
 
-        name = name.squeeze(' ') # spaces compressed
-        name = name.gsub /\A\s+/, '' # remove leading spaces
-        name = name.gsub /\s+\z/, '' # remove trailing spaces
+      def screening_hashes
+        screening_nodes.map { |node| ScreeningNodeParser.new(node).to_hash }
       end
 
-      # Showings
+      def screening_nodes
+        @node.css(SCREENING_CSS)
+      end
+
+      def tech
+        @tech ||= @node.css(TECH_CSS).text.gsub('in ', '').downcase
+      end
+    end
+
+    # parses a single screening
+    class ScreeningNodeParser
+      # regex for time format
+      TIME_REGEX = %r(\d+/\d+/\d+ \d{2}\:\d{2})
+      # regex for D-Box screenings
+      DBOX_REGEX = /D-Box/
+
+      # @param [Nokigiri::Node] node a Nokogiri node object
+      def initialize(node)
+        @node = node
+      end
+
+      # hashes of screening attributes
       # @return [Hash]
-      # @example
-      #   {
-      #     "2D" => [[Time.utc, 'http://www...'], [Time.utc, 'http://www...']],
-      #     "3D" => [[Time.utc, 'http://www...']]
-      #   }
-      def showings
-        tz = TZInfo::Timezone.get('Europe/London')
-        @nokogiri_html.css('.times-all.accordion-group').inject({}) do |result, variant_node|
-          variant = variant_node.css('.tech a').text.gsub('in ', '').upcase
-
-          times_url = variant_node.css('.performance-detail').map do |screening_node|
-            [
-              tz.local_to_utc(Time.parse(screening_node['title'].match(/\d+\/\d+\/\d+ \d{2}\:\d{2}/).to_s + ' UTC')),
-              "http://www.odeon.co.uk#{screening_node['href']}"
-            ]
-          end
+      def to_hash
+        {
+          booking_url: booking_url,
+          time:        utc_time,
+          variant:     variant
+        }
+      end
 
-          result.merge(variant => times_url)
-        end
+      private
+
+      def booking_url
+        link_attr('href').to_s
+      end
+
+      def link_attr(attribute)
+        @node.css('a').attribute(attribute)
+      end
+
+      def info
+        @node.css('i').to_s
+      end
+
+      def time
+        Time.parse(time_string)
+      end
+
+      def time_string
+        link_attr('title').to_s.match(TIME_REGEX).to_s + ' UTC'
+      end
+
+      def tz
+        @tz ||= TZInfo::Timezone.get('Europe/London')
+      end
+
+      def utc_time
+        tz.local_to_utc(time)
+      end
+
+      def variant
+        info.match(DBOX_REGEX) ? 'd-box' : ''
       end
     end
   end

data/lib/odeon_uk/internal/showtimes_page.rb

@@ -0,0 +1,36 @@
+module OdeonUk
+  # Internal utility classes: Do not use
+  # @api private
+  module Internal
+    # Parses a chunk of HTML to derive showing data for a single films
+    class ShowtimesPage
+      # css selector for film html chunks
+      FILM_CSS = '.film-detail'
+
+      # @param [Integer] cinema_id cineworld cinema id
+      def initialize(cinema_id)
+        @cinema_id = cinema_id
+      end
+
+      # break up the showtimes page into individual chunks for each film
+      # @return [Array<String>] html chunks for a film and it's screenings
+      def to_a
+        film_nodes.map { |node| node.to_s.gsub(/^\s+/, '') }
+      end
+
+      private
+
+      def film_nodes
+        @film_nodes ||= showtimes_doc.css(FILM_CSS)
+      end
+
+      def showtimes
+        @showtimes ||= OdeonUk::Internal::Website.new.showtimes(@cinema_id)
+      end
+
+      def showtimes_doc
+        @showtimes_doc ||= Nokogiri::HTML(showtimes)
+      end
+    end
+  end
+end

data/lib/odeon_uk/internal/title_sanitizer.rb

@@ -0,0 +1,49 @@
+module OdeonUk
+  # Internal utility classes: Do not use
+  # @api private
+  module Internal
+    # Sanitize and standardize film titles
+    class TitleSanitizer
+      # strings and regex to be removed
+      REMOVE = [
+        /\s+[23][dD]/,                 # dimension
+        'Autism Friendly Screening -', # autism screening
+        /\ACinemagic \d{1,4} \-/,      # cinemagic
+        /\(encore.+\)/i,               # encore for NT Live
+        /\(live\)/i,                   # live in brackets
+        'UKJFF -',                     # UK Jewish festival prefix
+      ]
+
+      # regexes and their replacements
+      REPLACE = {
+        /Bolshoi - (.*)/               => 'Bolshoi: ',
+        /Globe On Screen: (.*)/        => 'Globe: ',
+        /Met Opera - (.*)/             => 'Met Opera: ',
+        /National Theatre Live - (.*)/ => 'National Theatre: ',
+        /NT Live - (.*)/               => 'National Theatre: ',
+        /ROH - (.*)/                   => 'Royal Opera House: ',
+        /(.*)\- RSC Live \d{1,4}/      => 'Royal Shakespeare Company: '
+      }
+
+      # @param [String] title a film title
+      def initialize(title)
+        @title = title
+      end
+
+      # sanitized and standardized title
+      # @return [String] title
+      def sanitized
+        @sanitzed ||= begin
+          sanitized = @title
+          REMOVE.each do |pattern|
+            sanitized.gsub! pattern, ''
+          end
+          REPLACE.each do |pattern, prefix|
+            sanitized.gsub!(pattern) { |_| prefix + $1 }
+          end
+          sanitized.squeeze(' ').strip
+        end
+      end
+    end
+  end
+end

data/lib/odeon_uk/internal/website.rb

@@ -0,0 +1,36 @@
+require 'open-uri'
+
+module OdeonUk
+  # Internal utility classes: Do not use
+  # @api private
+  module Internal
+    # Utility class to make calls to the odeon website
+    class Website
+      # cinema page
+      # @param [Integer] cinema_id website id of the cinema
+      # @return [String] html of the page
+      def cinema(cinema_id)
+        get("cinemas/odeon/#{cinema_id}/")
+      end
+
+      # showtimes page for a single cinema
+      # @param [Integer] cinema_id website id of the cinema
+      # @return [String] html of the page
+      def showtimes(cinema_id)
+        get("showtimes/week/#{cinema_id}/?siteId=#{cinema_id}")
+      end
+
+      # sitemap page
+      # @return [String] html of the page
+      def sitemap
+        get('sitemap/')
+      end
+
+      private
+
+      def get(path)
+        URI("http://www.odeon.co.uk/#{path}").read
+      end
+    end
+  end
+end