cineworld_uk 1.0.5 → 2.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 0c228d909dc018b2ffa88f58841c53df17355f4c
-  data.tar.gz: 65837db94438ac24f9e8460f1765495835a482fc
+  metadata.gz: edcee9dcbbce90d73acbbb0ccc5650811bb7f931
+  data.tar.gz: 8454e532aa948af962692ce9d1add716b79dd405
 SHA512:
-  metadata.gz: 02646d623f4f7d13a8f3927283be7d17a9dbe0e8f8e2e3c3d1b31d40075858e28b6fb7b2aae321b72b3c1a25ab608f3821369bb2f608ebde31b75d1baa798130
-  data.tar.gz: 6cabf2784246c4d1816537221abd885ba19bbbca46335a6d5b1830e7da23e99914ae057e46c13d4182184fa0c6330fc07da06a3ed9f45578f231e898cac0a7d6
+  metadata.gz: 55369938b4df8ad874982a708dccb985b48f1514d2a6210e24371bcdf66f575dde627456cb4567fda757d37759bb9b0d7d6cccb8e2ed1865e2f70d7134ebefcf
+  data.tar.gz: 9d79dc209c5091269612dc732388d5160876defdaeef29ec7820656e5c1c3f32203144194dd5cfbc44b85ef5a1c041456a2a7974ff132941e79e9a894f6a3fd7

data/.travis.yml

@@ -2,4 +2,4 @@ language: ruby
 rvm:
   - 1.9.3
   - 2.0.0
-  - 2.1.0
+  - 2.1.2

data/CHANGELOG.md

@@ -1,3 +1,67 @@
+## 2.0.0 _25th July 2014_
+
+Bugfix
+
+- Cineworld changed their markup
+
+Features
+
+- Less logic in the Cinema class. Moved to `Film` and `Screening`.
+- Introduce `Internal::Website` class for all website comms
+- Rewrite Parsers and tests to be less exact but less fragile
+- Screening now uses the `FilmWithScreeningsParser`
+- Build `WhatsonParser` to break into 'films with screenings'
+- `FilmWithScreenings` then uses `ScreeningParser` class to get individual screening info
+- `ruby test/fixture_updator.rb` produces new fixtures from the live website
+  - it will break a few tests when run and committed but most should be ok
+- documentation
+
+Under the hood
+
+- Remove dependancy on HTTParty
+- Travis gets Ruby 2.1.2
+- General Rubocop improvements
+
+## 1.0.6 _28th Feb 2014_
+
+Bugfix
+
+- UTF encoding in Titleize internal class for ruby 1.9.3
+
+## 1.0.5 _28th Feb 2014_
+
+Bugfix
+
+- Vestigal require of titleize gem
+
+## 1.0.4 _27th Feb 2014_
+
+Under the hood
+
+- Standardizing film names using modified titlecase from `titlecase` gem
+- Film name parsing now moved to seperate class
+- rake console task added
+- MRI 2.1.0 support on Travis CI
+
+## 1.0.3, _16th Jan 2014_
+
+Additional methods
+
+- cinema#full_name including cinema brand
+- tidy up cinema naming to use colons
+
+## 1.0.2, _3rd Jan 2014_
+
+Under the hood
+
+- deal with HFR screenings, bloody hobbitses
+
+## 1.0.1, _3rd Jan 2014_
+
+External interface
+
+- screening#variant for spelling win, #varient maintained for compatability
+
 ## 1.0.0, _6th Dec 2013_
 
 First ready-for-public release

data/README.md

@@ -2,9 +2,10 @@
 
 A simple gem to parse the [Cineworld UK website](http://cineworld.co.uk) and spit out useful formatted info.
 
-[![Gem Version](https://badge.fury.io/rb/cineword_uk.png)](http://badge.fury.io/rb/cineworld_uk)
+[![Gem Version](https://badge.fury.io/rb/cineworld_uk.svg)](http://badge.fury.io/rb/cineworld_uk)
 [![Code Climate](https://codeclimate.com/github/andycroll/cineworld_uk.png)](https://codeclimate.com/github/andycroll/cineworld_uk)
 [![Build Status](https://travis-ci.org/andycroll/cineworld_uk.png?branch=master)](https://travis-ci.org/andycroll/cineworld_uk)
+[![Inline docs](http://inch-ci.org/github/andycroll/cineworld_uk.png)](http://inch-ci.org/github/andycroll/cineworld_uk)
 
 ## Installation
 

data/cineworld_uk.gemspec

@@ -18,11 +18,11 @@ Gem::Specification.new do |spec|
   spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
   spec.require_paths = ["lib"]
 
-  spec.add_development_dependency "bundler", "~> 1.3"
-  spec.add_development_dependency "rake"
+  spec.add_development_dependency 'bundler', '~> 1.3'
+  spec.add_development_dependency 'minitest-reporters'
+  spec.add_development_dependency 'rake'
   spec.add_development_dependency 'webmock'
 
-  spec.add_runtime_dependency 'httparty'
   spec.add_runtime_dependency 'nokogiri'
   spec.add_runtime_dependency 'tzinfo'
   spec.add_runtime_dependency 'tzinfo-data'

data/lib/cineworld_uk/cinema.rb

@@ -1,11 +1,9 @@
 module CineworldUk
-
   # The object representing a cinema on the Cineworld 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
+    # @return [Integer] the numeric id of the cinema on the Cineworld website
     attr_reader :id
     # @return [String] the name of the cinema
     attr_reader :name
@@ -20,8 +18,8 @@ module CineworldUk
     def initialize(id, name)
       @brand = 'Cineworld'
       @id    = id.to_i
-      @name  = name.gsub('London - ','').gsub(' - ', ': ')
-      @slug  = @name.downcase.gsub(/[^0-9a-z ]/,'').gsub(/\s+/, '-')
+      @name  = name.gsub('London - ', '').gsub(' - ', ': ')
+      @slug  = @name.downcase.gsub(/[^0-9a-z ]/, '').gsub(/\s+/, '-')
       @url   = "http://www.cineworld.co.uk/cinemas/#{@id}/information"
     end
 
@@ -29,32 +27,39 @@ module CineworldUk
     # @return [Array<CineworldUk::Cinema>]
     # @example
     #   CineworldUk::Cinema.all
-    #   #=> [<CineworldUk::Cinema brand="Cineworld" name="Brighton" slug="brighton" id=3 url="...">, <CineworldUk::Cinema brand="Cineworld" name="The O2, Greenwich" slug="the-o2-greenwich" url="...">, ...]
+    #   #=> [<CineworldUk::Cinema>, <CineworldUk::Cinema>, ...]
     def self.all
-      parsed_cinemas.css('#cinemaId option[value]').map do |option|
+      cinemas_doc.css('#cinemaId option[value]').map do |option|
         new option['value'], option.text
       end[1..-1]
     end
 
     # Find a single cinema
-    # @param [Integer, String] id the cinema id of the format 3/'3' as used on the cineworld.co.uk website
+    # @param [Integer, String] id the cinema id as used on the website
     # @return [CineworldUk::Cinema, nil]
     # @example
     #   CineworldUk::Cinema.find('3')
-    #   #=> <CineworldUk::Cinema brand="Cineworld" name="Brighton" slug="brighton" id=3 url="...">
+    #   #=> <CineworldUk::Cinema brand="Cineworld"
+    #                            name="Brighton"
+    #                            slug="brighton"
+    #                            id=3
+    #                            url="...">
     def self.find(id)
-      id = id.to_i
-      return nil unless id > 0
-
-      all.select { |cinema| cinema.id == id }[0]
+      all.select { |cinema| cinema.id == id.to_i }[0]
     end
 
     # Address of the cinema
     # @return [Hash] of different address parts
     # @example
-    #   cinema = CineworldUk::Cinema.find('Dukes_At_Komedia')
+    #   cinema = CineworldUk::Cinema.find(3)
     #   cinema.adr
-    #   #=> { street_address: '44-47 Gardner Street', extended_address: 'North Laine', locality: 'Brighton', postal_code: 'BN1 1UN', country_name: 'United Kingdom' }
+    #   #=> {
+    #         street_address: '44-47 Gardner Street',
+    #         extended_address: 'North Laine',
+    #         locality: 'Brighton',
+    #         postal_code: 'BN1 1UN',
+    #         country_name: 'United Kingdom'
+    #       }
     # @note Uses the standard method naming as at http://microformats.org/wiki/adr
     def adr
       {
@@ -80,20 +85,17 @@ module CineworldUk
     #   #=> nil
     # @note Uses the standard method naming as at http://microformats.org/wiki/adr
     def extended_address
-       remaining_address * ', ' unless remaining_address.empty?
+      remaining_address * ', ' unless remaining_address.empty?
     end
 
     # Films with showings scheduled at this cinema
-    # @return [Array<OdeonUk::Film>]
+    # @return [Array<CineworldUk::Film>]
     # @example
-    #   cinema = OdeonUk::Cinema.find('71')
+    #   cinema = CineworldUk::Cinema.find('71')
     #   cinema.films
-    #   #=> [<OdeonUk::Film name="Iron Man 3">, <OdeonUk::Film name="Star Trek Into Darkness">]
+    #   #=> [<CineworldUk::Film>, <CineworldUk::Film>, ...]
     def films
-      film_nodes.map do |node|
-        parser = CineworldUk::Internal::FilmWithScreeningsParser.new node.to_s
-        parser.film_name.length > 0 ? CineworldUk::Film.new(parser.film_name) : nil
-      end.compact.uniq
+      Film.at(id)
     end
 
     # The name of the cinema including the brand
@@ -114,7 +116,11 @@ module CineworldUk
     #   #=> 'Brighton'
     # @note Uses the standard method naming as at http://microformats.org/wiki/adr
     def locality
-      adr_has_region? && !adr_in_london? ? final_address_array[-2] : final_address_array[-1]
+      if adr_has_region? && !adr_in_london?
+        adr_array[-2]
+      else
+        adr_array[-1]
+      end
     end
 
     # Post code of the cinema
@@ -125,7 +131,7 @@ module CineworldUk
     #   #=> 'BN2 5UF'
     # @note Uses the standard method naming as at http://microformats.org/wiki/adr
     def postal_code
-      final_line_array[-2..-1]*' '
+      last_adr_line_array[-2..-1] * ' '
     end
 
     # The region (county) of the cinema if provided
@@ -136,7 +142,7 @@ module CineworldUk
     #   #=> 'East Sussex'
     # @note Uses the standard method naming as at http://microformats.org/wiki/adr
     def region
-      final_line_non_postal_code if adr_has_region? && !adr_in_london?
+      last_adr_line_non_postal_code if adr_has_region? && !adr_in_london?
     end
 
     # All planned screenings
@@ -144,16 +150,9 @@ module CineworldUk
     # @example
     #   cinema = CineworldUk::Cinema.find(3)
     #   cinema.screenings
-    #   # => [<CineworldUk::Screening film_name="Iron Man 3" cinema_name="Brighton" when="..." variant="...">, <CineworldUk::Screening ...>]
+    #   # => [<CineworldUk::Screening>, <CineworldUk::Screening>, ...]
     def screenings
-      film_nodes.map do |node|
-        parser = CineworldUk::Internal::FilmWithScreeningsParser.new node.to_s
-        parser.showings.map do |screening_type, times|
-          times.map do |array|
-            CineworldUk::Screening.new parser.film_name, self.name, array[0], array[1], screening_type
-          end
-        end
-      end.flatten
+      Screening.at(id)
     end
 
     # The street adress of the cinema
@@ -164,65 +163,57 @@ module CineworldUk
     #   #=> 'Brighton Marina'
     # @note Uses the standard method naming as at http://microformats.org/wiki/adr
     def street_address
-      address_parts[0]
+      adr_parts[0]
     end
 
     private
 
-    def self.parsed_cinemas
-      @parsed_cinemas ||= Nokogiri::HTML(cinemas_response)
-    end
-
-    def self.cinemas_response
-      @cinemas_response ||= HTTParty.get('http://www.cineworld.co.uk/cinemas')
-    end
-
-    def address_parts
-      @address_parts ||= parsed_information.css('address.marker').to_s.split('<br>')[-2].split(',').map(&:strip)
+    def self.cinemas_doc
+      @cinemas_doc ||= Nokogiri::HTML(website.cinemas)
     end
 
-    def adr_in_london?
-      final_line_non_postal_code == 'London'
+    def self.website
+      @website ||= CineworldUk::Internal::Website.new
     end
 
-    def adr_has_region?
-      final_line_non_postal_code != name
+    def adr_array
+      adr_parts[0..-2] << last_adr_line_non_postal_code
     end
 
-    def film_nodes
-      @film_nodes ||= parsed_whatson.css('.section.light #filter-reload > .span9').to_s.split('<hr>')
+    def adr_content
+      information_doc.css('address.marker').to_s
     end
 
-    def final_address_array
-      address_parts[0..-2] << final_line_non_postal_code
+    def adr_parts
+      @parts ||= adr_content.split('<br>')[-2].split(',').map(&:strip)
     end
 
-    def final_line_array
-      address_parts[-1].split(' ')
+    def adr_in_london?
+      last_adr_line_non_postal_code == 'London'
     end
 
-    def final_line_non_postal_code
-      final_line_array[0..-3]*' '
+    def adr_has_region?
+      last_adr_line_non_postal_code != name
     end
 
-    def information_response
-      @information_response ||= HTTParty.get("http://www.cineworld.co.uk/cinemas/#{@id}/information")
+    def last_adr_line_array
+      adr_parts[-1].split(' ')
     end
 
-    def parsed_information
-      @parsed_information ||= Nokogiri::HTML(information_response)
+    def last_adr_line_non_postal_code
+      last_adr_line_array[0..-3] * ' '
     end
 
-    def parsed_whatson
-      @parsed_whatson ||= Nokogiri::HTML(whatson_response)
+    def information_doc
+      @information_doc ||= begin
+        Nokogiri::HTML(CineworldUk::Internal::Website.new.information(@id))
+      end
     end
 
     def remaining_address
-      final_address_array.delete_if { |e| e == street_address || e == locality || e == region }
-    end
-
-    def whatson_response
-      @whatson_response ||= HTTParty.get("http://www.cineworld.co.uk/whatson?cinema=#{@id}")
+      adr_array.delete_if do |e|
+        e == street_address || e == locality || e == region
+      end
     end
   end
 end

data/lib/cineworld_uk/film.rb

@@ -1,5 +1,4 @@
 module CineworldUk
-
   # The object representing a film on the Cineworld UK website
   class Film
     include Comparable
@@ -13,14 +12,23 @@ module CineworldUk
     # @return [CineworldUk::Film]
     def initialize(name)
       @name = name
-      @slug = name.downcase.gsub(/[^0-9a-z ]/,'').gsub(/\s+/, '-')
+      @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<CineworldUk::Film>]
+    def self.at(cinema_id)
+      whatson_parser(cinema_id).films_with_screenings.map do |html|
+        new(screenings_parser(html).film_name)
+      end
     end
 
     # Allows sort on objects
     # @param [CineworldUk::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.
@@ -28,20 +36,30 @@ module CineworldUk
     # @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
 
     # Generates hash of slug in order to allow two records of the same type and
     # id to work with something like:
     #
-    #   [ Film.new('ABC'), Film.new('DEF') ] & [ Film.new('DEF'), Film.new('GHI') ]
+    #   [Film.new('ABC'), Film.new('DEF')] & [Film.new('DEF'), Film.new('GHI')]
     #   #=> [ Film.new('DEF') ]
     #
     # @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)
+      CineworldUk::Internal::FilmWithScreeningsParser.new(html)
+    end
+
+    def self.whatson_parser(id)
+      CineworldUk::Internal::WhatsonParser.new(id)
     end
   end
 end

data/lib/cineworld_uk/internal/film_with_screenings_parser.rb

@@ -1,102 +1,60 @@
 module CineworldUk
-
   # 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 film name
+      FILM_NAME_CSS    = 'h3.h1 a[href*=whatson]'
+      # css selector for performances
+      PERFORMANCES_CSS = '.schedule .performances > li'
 
       # @param [String] film_html a chunk of html
       def initialize(film_html)
-        @nokogiri_html = Nokogiri::HTML(film_html)
+        @film_html = film_html.to_s
+      end
+
+      # The cinema id
+      # @return [String]
+      def cinema_id
+        name_doc.to_s.match(/cinema=(\d+)/)[1].to_i
       end
 
       # The film name
       # @return [String]
       def film_name
-        NameParser.new(original_name).standardize
+        name_doc.children[0].to_s
       end
 
-      # Showings
-      # @return [Hash]
-      # @example
-      #   {
-      #     "2D" => [Time.utc, Time.utc]
-      #   }
-      def showings
-        tz = TZInfo::Timezone.get('Europe/London')
-        @nokogiri_html.css('.schedule .performances > li').inject({}) do |result, li|
-          key = performance_variant(li)
-
-          if has_bookable_link_node?(li)
-            time_array = performance_date_array(li) + performance_time_array(li)
-            time = tz.local_to_utc(Time.utc(*time_array))
-            result.merge(key => (result[key] || []) << [time, booking_url(li)])
-          else
-            result
-          end
-        end
+      # attributes of all the screenings
+      # @return [Array<Hash>]
+      def to_a
+        performances_doc.map do |node|
+          next unless screening_parser_hash(node)
+          screening_parser_hash(node).merge(film_hash)
+        end.compact
       end
 
       private
 
-      def booking_url(node)
-        'http://www.cineworld.co.uk' + performance_link(node)['href']
-      end
-
-      def dbox?(node)
-        node.css('.tooltip-box .icon-service-dbox').count > 0
-      end
-
-      def dimension(node)
-        node.css('.tooltip-box .icon-service-twod, .tooltip-box .icon-service-thrd').text
-      end
-
-      def has_bookable_link_node?(node)
-        performance_link(node) != nil
-      end
-
-      def imax?(node)
-        node.css('.tooltip-box .icon-service-imax').count > 0
-      end
-
-      def hfr?(node)
-        node.css('.tooltip-box .icon-service-hfr').count > 0
-      end
-
-      def original_name
-        @original_name ||= @nokogiri_html.css('.span5 h1 a,.span7 h1 a, .span7 h1').children[0].to_s
-      end
-
-      def performance_date_array(node)
-        if has_bookable_link_node?(node)
-          match = performance_link_text(node).match(/date\=(\d{4})(\d{2})(\d{2})/)
-          [match[1], match[2], match[3]]
-        end
-      end
-
-      def performance_link(node)
-        node.css('a.performance').first
+      def doc
+        @doc ||= Nokogiri::HTML(@film_html)
       end
 
-      def performance_link_string(node)
-        performance_link(node).to_s
+      def film_hash
+        @film_hash ||= { cinema_id: cinema_id, film_name: film_name }
       end
 
-      def performance_link_text(node)
-        has_bookable_link_node?(node) ? performance_link_string(node) : nil
+      def name_doc
+        @name_doc ||= doc.css(FILM_NAME_CSS)
       end
 
-      def performance_time_array(node)
-        if has_bookable_link_node?(node)
-          match = performance_link_text(node).match(/time\=(\d{2})\:(\d{2})/)
-          [match[1], match[2]]
-        end
+      def performances_doc
+        @performances_doc ||= doc.css(PERFORMANCES_CSS)
       end
 
-      def performance_variant(node)
-        dimension(node) + "#{ ' D-BOX' if dbox?(node) }#{ ' IMAX' if imax?(node) }#{ ' HFR' if hfr?(node) }"
+      def screening_parser_hash(node)
+        ScreeningParser.new(node).to_hash
       end
     end
   end

data/lib/cineworld_uk/internal/name_parser.rb

@@ -1,14 +1,13 @@
 module CineworldUk
-
   # Internal utility classes: Do not use
   # @api private
   module Internal
-
     # Parses a string to derive a standardized movie title
     class NameParser
-
+      # @return [String] the original name
       attr_reader :original_name
 
+      # @param [String] name original film name
       def initialize(name)
         @original_name = name
         @name = name
@@ -17,15 +16,15 @@ module CineworldUk
       # Process the name and return the final string
       # @return [String]
       def standardize
-        strip_and_squeeze.
-          ampersands_into_text.
-          into_ampersand_if_second_to_last.
-          remove_indian_languages.
-          remove_screening_details.
-          replace_non_film_prefix.
-          remove_newlines.
-          remove_dates.
-          title_case
+        strip_and_squeeze
+          .ampersands_into_text
+          .into_ampersand_if_second_to_last
+          .remove_indian_languages
+          .remove_screening_details
+          .replace_non_film_prefix
+          .remove_newlines
+          .remove_dates
+          .title_case
         to_s
       end
 
@@ -48,9 +47,9 @@ module CineworldUk
       end
 
       def remove_indian_languages
-        languages = ['Malayalam', 'Tamil']
+        languages = %w(Malayalam Tamil)
 
-        _remove(/\((#{languages*'|'})\)/i)
+        _remove(/\((#{languages * '|'})\)/i)
         self
       end
 
@@ -61,7 +60,7 @@ module CineworldUk
       end
 
       def remove_dates
-        _remove(/\-? \d{1,2}\/\d{1,2}\/\d{2,4}/)
+        _remove(%r(-? \d{1,2}/\d{1,2}/\d{2,4}))
         self
       end
 
@@ -73,15 +72,18 @@ module CineworldUk
       def replace_non_film_prefix
         _replace 'Bolshoi Ballet Live -', 'Bolshoi:'
 
-        @name = 'National Theatre: ' + @name.gsub(/\- NT .+ encore/, '') if @name.match /\- NT .+ encore/
+        if @name.match(/\- NT .+ encore/)
+          @name = 'National Theatre: ' + @name.gsub(/\- NT .+ encore/, '')
+        end
+
         _replace 'NT Live:', 'National Theatre:'
 
         _replace 'MET Opera -', 'Met Opera:'
         _replace 'Royal Ballet Live:', 'Royal Ballet:'
 
         # fill out Royal Opera House
-        if pure_name_match = @name.match(/Royal Opera Live\: (.+) \-.+/)
-          @name = 'Royal Opera House: ' + pure_name_match[1]
+        @name.match(/Royal Opera Live\: (.+) \-.+/) do |match|
+          @name = 'Royal Opera House: ' + match[1]
         end
         _replace 'Royal Opera Live:', 'Royal Opera House:'