article_json 0.3.8 → 0.4.1

data/lib/article_json/export/apple_news/elements/paragraph.rb

@@ -0,0 +1,36 @@
+module ArticleJSON
+  module Export
+    module AppleNews
+      module Elements
+        class Paragraph < Base
+          # Generate the paragraph node with its containing text elements
+          # @return [Hash]
+          def export
+            {
+              role: 'body',
+              text: text,
+              format: 'html',
+              layout: 'bodyLayout',
+              textStyle: 'bodyStyle',
+            }
+          end
+
+          private
+
+          # Get the exporter class for text elements
+          # @return [ArticleJSON::Export::Common::HTML::Elements::Base]
+          def text_exporter
+            self.class.exporter_by_type(:text)
+          end
+
+          def text
+            @element.content.map do |child_element|
+              text_exporter.new(child_element)
+                           .export
+            end.join
+          end
+        end
+      end
+    end
+  end
+end

data/lib/article_json/export/apple_news/elements/quote.rb

@@ -0,0 +1,60 @@
+module ArticleJSON
+  module Export
+    module AppleNews
+      module Elements
+        class Quote < Base
+          include ArticleJSON::Export::Common::HTML::Elements::Base
+          include ArticleJSON::Export::Common::HTML::Elements::Text
+
+          def export
+            [quote, author]
+          end
+
+          private
+
+          # Quote
+          # @return [Hash]
+          def quote
+            {
+              role: 'pullquote',
+              text: quote_text,
+              format: 'html',
+              layout: 'pullquoteLayout',
+              textStyle: 'pullquoteStyle',
+            }
+          end
+
+          # Author
+          # @return [Hash]
+          def author
+            {
+              role: 'author',
+              text: author_text,
+              format: 'html',
+              layout: 'pullquoteAttributeLayout',
+              textStyle: 'quoteAttributeStyle',
+            }
+          end
+
+          def text_exporter
+            self.class.exporter_by_type(:text)
+          end
+
+          # Quote Text
+          # @return [String]
+          def quote_text
+            element = @element.content.first&.content.first
+            text_exporter.new(element).export
+          end
+
+          # Author Text
+          # @return [String]
+          def author_text
+            element = @element.caption.first
+            text_exporter.new(element).export
+          end
+        end
+      end
+    end
+  end
+end

data/lib/article_json/export/apple_news/elements/text.rb

@@ -0,0 +1,55 @@
+module ArticleJSON
+  module Export
+    module AppleNews
+      module Elements
+        class Text < Base
+          include ArticleJSON::Export::Common::HTML::Elements::Base
+          include ArticleJSON::Export::Common::HTML::Elements::Text
+
+          UNSUPPORTED_HTML_TAGS = %w[
+            title
+            meta
+            script
+            noscript
+            style
+            link
+            applet
+            object
+            iframe
+            noframes
+            form
+            select
+            option
+            optgroup
+          ].freeze
+
+          # A Nokogiri object is returned with`super`, which is is then
+          # returned as a either a string or as HTML (when not plain text),
+          # both of which are compatible with Apple News format. Takes into
+          # account bold, italic and href.
+          # @return [String]
+          def export
+            super.to_s
+          end
+
+          # @param [String] text
+          def create_text_nodes(text)
+            Nokogiri::HTML.fragment(sanitize_text(text).gsub(/\n/, '<br>')).children
+          end
+
+          # Removes UNSUPPORTED_TAGS from text
+          #
+          # @param [String] text
+          # @return [String]
+          def sanitize_text(text)
+            doc = Nokogiri::HTML.fragment(text)
+            UNSUPPORTED_HTML_TAGS.each do |tag|
+              doc.search(tag).each(&:remove)
+            end
+            doc.inner_html
+          end
+        end
+      end
+    end
+  end
+end

data/lib/article_json/export/apple_news/elements/text_box.rb

@@ -0,0 +1,51 @@
+module ArticleJSON
+  module Export
+    module AppleNews
+      module Elements
+        class TextBox < Base
+          include ArticleJSON::Export::Common::HTML::Elements::TextBox
+          # List
+          # @return [Hash]
+          def export
+            {
+              role: 'container',
+              layout: 'textBoxLayout',
+              style: 'textBoxStyle',
+              components: map_styles(elements),
+            }
+          end
+
+          private
+
+          # @return [Array]
+          def elements
+            @element.content.map do |child_element|
+              case child_element
+              when ArticleJSON::Elements::Heading
+                namespace::Heading.new(child_element).export
+              when ArticleJSON::Elements::Paragraph
+                namespace::Paragraph.new(child_element).export
+              when ArticleJSON::Elements::List
+                namespace::List.new(child_element).export
+              else
+                namespace::Text.new(child_element).export
+              end
+            end
+          end
+
+          # @return [Module]
+          def namespace
+            ArticleJSON::Export::AppleNews::Elements
+          end
+
+          # @return [Array]
+          def map_styles(elements)
+            elements.map do |child_element|
+              child_element.merge(layout: 'textBox' +  child_element[:layout].sub(/\S/, &:upcase))
+            end
+          end
+        end
+      end
+    end
+  end
+end

data/lib/article_json/export/apple_news/exporter.rb

@@ -0,0 +1,37 @@
+module ArticleJSON
+  module Export
+    module AppleNews
+      class Exporter
+        # @param [Array[ArticleJSON::Elements::Base]] elements
+        def initialize(elements)
+          @elements = elements
+        end
+
+        # Return the components section of an Apple News Article as JSON
+        #
+        # Images and EmbededVideos are nested in an array with the components
+        # array when they contain captions. As Apple News skips over these
+        # nested arrays, we must flatten the array.
+        #
+        # @return [String]
+        def to_json
+          { components: components.flatten }.to_json
+        end
+
+        private
+
+        # Generate an array with the plain text representation of all elements
+        #
+        # @return [Array]
+        def components
+          @components ||=
+            @elements.map do |element|
+              ArticleJSON::Export::AppleNews::Elements::Base
+                .build(element)
+                &.export
+            end.reject { |hash| hash.nil? || hash.empty? }
+        end
+      end
+    end
+  end
+end

data/lib/article_json/export/common/html/elements/embed.rb

@@ -20,7 +20,7 @@ module ArticleJSON
             private
 
             def embed_node
-              type = @element.embed_type.to_s.tr('_','-')
+              type = @element.embed_type.to_s.tr('_', '-')
               create_element(:div, class: "embed #{type}") do |div|
                 div.add_child(embedded_object)
               end
@@ -28,6 +28,7 @@ module ArticleJSON
 
             def embedded_object
               return unavailable_node unless @element.oembed_data
+
               Nokogiri::HTML.fragment(@element.oembed_data[:html])
             end
 

data/lib/article_json/export/common/html/elements/image.rb

@@ -19,7 +19,7 @@ module ArticleJSON
             # @return [Nokogiri::XML::NodeSet]
             def figure_node
               create_element(:figure, node_opts) do |figure|
-                node =  @element&.href ? href_node : image_node
+                node = @element&.href ? href_node : image_node
                 figure.add_child(node)
                 if @element.caption&.any?
                   figure.add_child(caption_node(:figcaption))
@@ -42,6 +42,7 @@ module ArticleJSON
             # @return [Hash]
             def node_opts
               return if floating_class.nil?
+
               { class: floating_class }
             end
           end

data/lib/article_json/export/common/html/elements/text.rb

@@ -11,6 +11,7 @@ module ArticleJSON
               return bold_and_italic_node if @element.bold && @element.italic
               return bold_node if @element.bold
               return italic_node if @element.italic
+
               content_node
             end
 
@@ -38,6 +39,7 @@ module ArticleJSON
             # @return [Nokogiri::XML::NodeSet]
             def content_node
               return create_text_nodes(@element.content) if @element.href.nil?
+
               create_element(:a, href: @element.href) do |a|
                 a.add_child(create_text_nodes(@element.content))
               end

data/lib/article_json/import/google_doc/html/embedded_parser.rb

@@ -104,6 +104,7 @@ module ArticleJSON
             def find_parser(text)
               text = text.strip.downcase
               return nil if text.empty?
+
               parsers.find { |klass| klass.matches?(text) }
             end
           end

data/lib/article_json/import/google_doc/html/heading_parser.rb

@@ -19,11 +19,11 @@ module ArticleJSON
           # @return [Integer]
           def level
             case @node.name
-              when 'h1' then 1
-              when 'h2' then 2
-              when 'h3' then 3
-              when 'h4' then 4
-              when 'h5' then 5
+            when 'h1' then 1
+            when 'h2' then 2
+            when 'h3' then 3
+            when 'h4' then 4
+            when 'h5' then 5
             end
           end
 

data/lib/article_json/import/google_doc/html/image_parser.rb

@@ -22,25 +22,31 @@ module ArticleJSON
           # The value of the image's `alt` attribute
           # @return [String]
           def alt
+            return '' if image_url?
+
             image_node.attribute('alt')&.value || ''
           end
 
           # The value of the image's `src` attribute
           # @return [String]
           def source_url
+            return @node.inner_text.strip if image_url?
+
             image_node.attribute('src').value
           end
 
           # The node of the actual image
           # @return [Nokogiri::HTML::Node]
           def image_node
-            @node.xpath('.//img').first
+            return @image_node if defined? @image_node
+
+            @image_node = @node.xpath('.//img').first
           end
 
           # Check if the image is floating (left, right or not at all)
           # @return [Symbol]
           def float
-            super if floatable_size?
+            super if image_url? || floatable_size?
           end
 
           # Extracts an href from the tag [image-link-to: url]) if present
@@ -48,8 +54,10 @@ module ArticleJSON
           # @return [String]
           def href
             return if @caption_node.nil?
+
             match = @caption_node.content.strip.match(href_regexp)
             return if match.nil?
+
             remove_image_link_tag
             match[:url]
           end
@@ -80,6 +88,7 @@ module ArticleJSON
           def href_regexp
             %r{\[image-link-to:\s+(?<url>.*?)\]}
           end
+
           # Check if the image's width can be determined and is less than 500px
           # This is about 3/4 of the google document width...
           # @return [Boolean]
@@ -101,6 +110,13 @@ module ArticleJSON
                 match['px'].to_i if match && match['px']
               end
           end
+
+          # When the current node doesn't contain an actual image tag,
+          # we're dealing with an image URL
+          # @return [Boolean]
+          def image_url?
+            image_node.nil?
+          end
         end
       end
     end

data/lib/article_json/import/google_doc/html/list_parser.rb

@@ -14,8 +14,8 @@ module ArticleJSON
           # @return [Symbol]
           def list_type
             case @node.name
-              when 'ol' then :ordered
-              when 'ul' then :unordered
+            when 'ol' then :ordered
+            when 'ul' then :unordered
             end
           end
 

data/lib/article_json/import/google_doc/html/node_analyzer.rb

@@ -31,6 +31,7 @@ module ArticleJSON
           # @return [Boolean]
           def empty?
             return @is_empty if defined? @is_empty
+
             @is_empty = node.inner_text.strip.empty? && !image? && !hr? && !br?
           end
 
@@ -38,6 +39,7 @@ module ArticleJSON
           # @return [Boolean]
           def heading?
             return @is_heading if defined? @is_heading
+
             @is_heading =
               !quote? && !text_box? && %w(h1 h2 h3 h4 h5).include?(node.name)
           end
@@ -52,6 +54,7 @@ module ArticleJSON
           # @return [Boolean]
           def paragraph?
             return @is_paragraph if defined? @is_paragraph
+
             @is_paragraph =
               node.name == 'p' &&
                 !empty? &&
@@ -65,7 +68,8 @@ module ArticleJSON
           # @return [Boolean]
           def list?
             return @is_list if defined? @is_list
-            @is_list = %w(ul ol).include?(node.name)
+
+            @is_list = %w[ul ol].include?(node.name)
           end
 
           # Check if the node starts a text box
@@ -73,6 +77,7 @@ module ArticleJSON
           # @return [Boolean]
           def text_box?
             return @is_text_box if defined? @is_text_box
+
             @is_text_box = begins_with?('textbox:') || begins_with?('highlight:')
           end
 
@@ -81,6 +86,7 @@ module ArticleJSON
           # @return [Boolean]
           def quote?
             return @is_quote if defined? @is_quote
+
             @is_quote = has_text?('quote:')
           end
 
@@ -88,13 +94,25 @@ module ArticleJSON
           # @return [Boolean]
           def image?
             return @is_image if defined? @is_image
-            @is_image = node.xpath('.//img').length > 0
+
+            @is_image = image_url? || node.xpath('.//img').length > 0
+          end
+
+          # Check if the node contains an image URL
+          # @return [Boolean]
+          def image_url?
+            return @is_image_url if defined? @is_image_url
+
+            text = node.inner_text.strip
+            url_regexp = %r{https?:\/\/\S+\.(?:jpg|jpeg|png|gif)}i
+            @is_image_url = !!(url_regexp =~ text)
           end
 
           # Check if the node contains an embedded element
           # @return [Boolean]
           def embed?
             return @is_embed if defined? @is_embed
+
             @is_embed = EmbeddedParser.supported?(node)
           end
 
@@ -103,6 +121,7 @@ module ArticleJSON
           # @return [Boolean]
           def br?
             return @is_br if defined? @is_br
+
             @is_br = node.name == 'br' || only_includes_brs?
           end
 
@@ -119,6 +138,7 @@ module ArticleJSON
             return :quote if quote?
             return :image if image?
             return :embed if embed?
+
             :unknown
           end
 
@@ -128,9 +148,11 @@ module ArticleJSON
           # @return [Boolean]
           def only_includes_brs?
             return false unless node.inner_text.strip.empty?
+
             tags = node.children.map(&:name)
             # Check if it only contains <br> and text nodes
-            return false unless tags.all? { |tag| %w(br text).include? tag }
+            return false unless tags.all? { |tag| %w[br text].include? tag }
+
             # Check if at least one is a `<br>` node
             tags.include?('br')
           end

data/lib/article_json/import/google_doc/html/parser.rb

@@ -6,7 +6,12 @@ module ArticleJSON
           # @param [String] html
           def initialize(html)
             doc = Nokogiri::HTML(html)
-            @body_enumerator = doc.xpath('//body').last.children.to_enum
+            selection = if doc.xpath('//body/div').empty?
+                          doc.xpath('//body')
+                        else
+                          doc.xpath('//body/div')
+                        end
+            @body_enumerator = selection.last.children.to_enum
 
             css_node = doc.xpath('//head/style').last
             @css_analyzer = CSSAnalyzer.new(css_node&.inner_text)
@@ -113,6 +118,7 @@ module ArticleJSON
             nodes = []
             until !body_has_more_nodes? ||
                 NodeAnalyzer.new(@body_enumerator.peek).hr?
+
               nodes << @body_enumerator.next
             end
             nodes

data/lib/article_json/import/google_doc/html/shared/caption.rb

@@ -8,6 +8,7 @@ module ArticleJSON
             # @return [Array[ArticleJSON::Elements::Text]]
             def caption
               return [] if no_caption?
+
               ArticleJSON::Import::GoogleDoc::HTML::TextParser.extract(
                 node: @caption_node,
                 css_analyzer: @css_analyzer

data/lib/article_json/import/google_doc/html/shared/float.rb

@@ -8,9 +8,11 @@ module ArticleJSON
             # @return [Symbol]
             def float
               return unless @float_node.has_attribute?('class')
+
               node_class = @float_node.attribute('class').value || ''
               return :right if @css_analyzer.right_aligned?(node_class)
               return :left if @css_analyzer.left_aligned?(node_class)
+
               nil
             end
           end

data/lib/article_json/import/google_doc/html/text_box_parser.rb

@@ -10,7 +10,7 @@ module ArticleJSON
           #                                         May contain tags, too.
           # @param [Array[Nokogiri::HTML::Node]] nodes
           # @param [ArticleJSON::Import::GoogleDoc::HTML::CSSAnalyzer] css_analyzer
-          def initialize(type_node: ,nodes:, css_analyzer:)
+          def initialize(type_node:, nodes:, css_analyzer:)
             @nodes = nodes.reject { |node| NodeAnalyzer.new(node).empty? }
             @css_analyzer = css_analyzer
 
@@ -32,6 +32,7 @@ module ArticleJSON
             match = /(.*?)[\s\u00A0]+\[(?<tags>.*)\]/
                       .match(@type_node.inner_text)
             return [] unless match
+
             match[:tags].split(' ')
           end
 

data/lib/article_json/import/google_doc/html/text_parser.rb

@@ -43,6 +43,7 @@ module ArticleJSON
             if @node.name == 'span' &&
                 @node.first_element_child&.name == 'a' &&
                 @node.first_element_child&.has_attribute?('href')
+
               strip_google_redirect(
                 @node.first_element_child.attribute('href').value
               )
@@ -68,6 +69,7 @@ module ArticleJSON
             def extract(node:, css_analyzer:)
               node.children.map do |child_node|
                 next if NodeAnalyzer.new(child_node).empty?
+
                 new(node: child_node, css_analyzer: css_analyzer).element
               end.compact
             end

data/lib/article_json/utils/additional_element_placer.rb

@@ -40,6 +40,7 @@ module ArticleJSON
       # @return [Array[ArticleJSON::Elements::Base|Object]]
       def merge_elements
         return @additional_elements if @elements.nil? || @elements.empty?
+
         remaining_elements = @additional_elements.dup
         next_in = insert_next_element_in(0, remaining_elements)
         characters_passed = 0
@@ -48,6 +49,7 @@ module ArticleJSON
           .each_with_object([]) do |(element, next_element), result|
             result << element
             next if remaining_elements.empty?
+
             if element.respond_to?(:length)
               characters_passed += element.length
               next_in -= element.length

data/lib/article_json/utils/o_embed_resolver/base.rb

@@ -23,8 +23,10 @@ module ArticleJSON
         def unavailable_message
           [
             ArticleJSON::Elements::Text.new(content: "The #{name} "),
-            ArticleJSON::Elements::Text.new(content: source_url,
-                                            href: source_url),
+            ArticleJSON::Elements::Text.new(
+              content: source_url,
+              href: source_url
+            ),
             ArticleJSON::Elements::Text.new(content: ' is not available.'),
           ]
         end
@@ -44,19 +46,27 @@ module ArticleJSON
         # @return [Hash|nil]
         def parsed_api_response
           return @api_response if defined? @api_response
+
           @api_response = begin
             uri = URI.parse(oembed_url)
             http = Net::HTTP.new(uri.host, uri.port)
             http.use_ssl = (uri.scheme == 'https')
             response = http.request(Net::HTTP::Get.new(uri, http_headers))
-            if response.kind_of? Net::HTTPSuccess
-              JSON.parse(response.body, symbolize_names: true)
+
+            if response.is_a? Net::HTTPSuccess
+              data = JSON.parse(response.body, symbolize_names: true)
+              transform_api_response(data)
             end
           rescue Net::ProtocolError, JSON::ParserError
             nil
           end
         end
 
+        # @return [Hash]
+        def transform_api_response(data)
+          data
+        end
+
         # @return [Hash]
         def http_headers
           headers = { 'Content-Type' => 'application/json' }

data/lib/article_json/utils/o_embed_resolver/facebook_video.rb

@@ -11,7 +11,8 @@ module ArticleJSON
         # The URL for the oembed API call
         # @return [String]
         def oembed_url
-          "https://www.facebook.com/plugins/video/oembed.json?url=#{source_url}"
+          "https://graph.facebook.com/v9.0/oembed_video?url=#{source_url}" \
+            "&access_token=#{access_token}"
         end
 
         # The video URL of the element
@@ -19,6 +20,21 @@ module ArticleJSON
         def source_url
           "https://www.facebook.com/facebook/videos/#{@element.embed_id}"
         end
+
+        # The facebook access token. If not set, it raises an exception
+        # explaining how to configure it.
+        #
+        # @return [String]
+        def access_token
+          token = ArticleJSON.configuration.facebook_token
+
+          if token.nil?
+            raise 'You need to configure the facebook token to use facebook' \
+                  'embed videos, see:' \
+                  'https://github.com/Devex/article_json#facebook-oembed'
+          end
+          token
+        end
       end
     end
   end