microformats 4.0.6 → 4.2.1

data/lib/microformats/parser.rb

@@ -1,14 +1,13 @@
 module Microformats
-
   class Parser < ParserCore
     attr_reader :http_headers, :http_body
+
     def initialize
       @http_headers = {}
       super
     end
 
-    def parse(html, base: nil, headers:{})
-
+    def parse(html, base: nil, headers: {})
       @http_headers = {}
 
       @items = []
@@ -26,49 +25,51 @@ module Microformats
       @base = found_base unless found_base.nil?
 
       document.traverse do |node|
-        if not node.attribute('src').nil?
+        if !node.attribute('src').nil?
           absolute_url = Microformats::AbsoluteUri.new(node.attribute('src').value.to_s, base: @base).absolutize
           node.attribute('src').value = absolute_url.to_s
-
-        elsif not node.attribute('href').nil?
+        elsif !node.attribute('href').nil?
           absolute_url = Microformats::AbsoluteUri.new(node.attribute('href').value.to_s, base: @base).absolutize
           node.attribute('href').value = absolute_url.to_s
         end
       end
+
       parse_node(document)
       parse_rels(document)
 
-      Collection.new({'items' => @items, 'rels' => @rels, 'rel-urls' =>  @rel_urls})
+      Collection.new('items' => @items, 'rels' => @rels, 'rel-urls' => @rel_urls)
     end
 
-    def read_html(html, headers:{})
+    def read_html(html, headers: {})
       stripped_html = html.strip
-      open(stripped_html, headers) do |response|
+
+      (URI.respond_to?(:open) ? URI : Kernel).open(stripped_html, **headers) do |response|
         @http_headers = response.meta if response.respond_to?(:meta)
         @http_body = response.read
       end
-      if @base.nil?
-          @base = stripped_html
-      end
+
+      @base = stripped_html if @base.nil?
+
       @http_body
-    rescue Errno::ENOENT, Errno::ENAMETOOLONG => e
+    rescue Errno::ENOENT, Errno::ENAMETOOLONG
       @http_body = html
     end
 
     private
 
     def parse_element(element)
-
       fmt_classes = format_classes(element)
+
       bc_fmt_classes = backcompat_format_classes(element).reject do |format_class|
-        fmt_classes.include? format_class
+        fmt_classes.include?(format_class)
       end
-      joined_classes =  fmt_classes + bc_fmt_classes
+
+      joined_classes = fmt_classes + bc_fmt_classes
 
       if bc_fmt_classes.length >= 1
         @items << FormatParser.new.parse(element, base: @base, format_class_array: joined_classes, backcompat: true)
       elsif fmt_classes.length >= 1
-        @items << FormatParser.new.parse(element, base: @base, format_class_array: fmt_classes )
+        @items << FormatParser.new.parse(element, base: @base, format_class_array: fmt_classes)
       else
         parse_nodeset(element.children)
       end
@@ -76,36 +77,35 @@ module Microformats
 
     def parse_base(document)
       base = document.search('base').first
+
       base.values.first unless base.nil?
     end
 
     def parse_rels(element)
       element.search('*[@rel]').each do |rel|
-        unless rel.attribute('href').nil?
-          url = Microformats::AbsoluteUri.new(rel.attribute('href').text, base: @base).absolutize
-
-          rel_values = rel.attribute('rel').text.split(' ')
-          rel_values.each do |rel_value|
-            @rels[rel_value] = [] unless @rels.has_key?(rel_value)
-            @rels[rel_value] << Microformats::AbsoluteUri.new(rel.attribute('href').text, base: @base).absolutize
-            @rels[rel_value].uniq!
-          end
-
-          unless rel_values.empty? 
-              @rel_urls[url] = {} unless @rel_urls.has_key?(url)
-
-              @rel_urls[url]['hreflang'] = rel.attribute('hreflang').value if @rel_urls[url]['hreflang'].nil? and not rel.attribute('hreflang').nil?
-              @rel_urls[url]['media'] = rel.attribute('media').value if @rel_urls[url]['media'].nil? and not rel.attribute('media').nil?
-              @rel_urls[url]['title'] = rel.attribute('title').value if @rel_urls[url]['title'].nil? and not rel.attribute('title').nil?
-              @rel_urls[url]['type'] = rel.attribute('type').value if @rel_urls[url]['type'].nil? and not rel.attribute('type').nil?
-              @rel_urls[url]['text'] = rel.text.strip if  @rel_urls[url]['text'].nil? and not rel.text.empty?
-              @rel_urls[url]['rels'] = rel_values
-          end
+        next if rel.attribute('href').nil?
+
+        url = Microformats::AbsoluteUri.new(rel.attribute('href').text, base: @base).absolutize
+
+        rel_values = rel.attribute('rel').text.split(' ')
 
+        rel_values.each do |rel_value|
+          @rels[rel_value] = [] unless @rels.key?(rel_value)
+          @rels[rel_value] << Microformats::AbsoluteUri.new(rel.attribute('href').text, base: @base).absolutize
+          @rels[rel_value].uniq!
         end
+
+        next if rel_values.empty?
+
+        @rel_urls[url] = {} unless @rel_urls.key?(url)
+
+        @rel_urls[url]['hreflang'] = rel.attribute('hreflang').value if @rel_urls[url]['hreflang'].nil? && !rel.attribute('hreflang').nil?
+        @rel_urls[url]['media'] = rel.attribute('media').value if @rel_urls[url]['media'].nil? && !rel.attribute('media').nil?
+        @rel_urls[url]['title'] = rel.attribute('title').value if @rel_urls[url]['title'].nil? && !rel.attribute('title').nil?
+        @rel_urls[url]['type'] = rel.attribute('type').value if @rel_urls[url]['type'].nil? && !rel.attribute('type').nil?
+        @rel_urls[url]['text'] = rel.text.strip if @rel_urls[url]['text'].nil? && !rel.text.empty?
+        @rel_urls[url]['rels'] = rel_values
       end
     end
-
   end
 end
-

data/lib/microformats/parser_core.rb

@@ -1,12 +1,10 @@
 module Microformats
   class ParserCore
-
     VALUE_CLASS_REG_EXP = /^value$/
     VALUE_TITLE_CLASS_REG_EXP = /^value-title$/
     FORMAT_CLASS_REG_EXP = /^h-[a-z-]+$/
     PROPERTY_CLASS_REG_EXP = /^(p-|u-|dt-|e-)[a-z-]+$/
 
-
     def initialize
       @mode_backcompat = false
       @fmt_classes = []
@@ -15,18 +13,15 @@ module Microformats
     private
 
     def parse_node(node)
-      if node.is_a?(Nokogiri::HTML::Document) then
+      if node.is_a?(Nokogiri::HTML::Document)
         parse_node(node.children)
-      elsif node.is_a?(Nokogiri::XML::NodeSet) then
+      elsif node.is_a?(Nokogiri::XML::NodeSet)
         parse_nodeset(node)
-      elsif node.is_a?(Nokogiri::XML::Element) then
+      elsif node.is_a?(Nokogiri::XML::Element)
         parse_element(node)
-      else
-        nil
       end
     end
 
-
     def parse_nodeset(nodeset)
       nodeset.each do |node|
         parse_node(node)
@@ -67,60 +62,71 @@ module Microformats
         when  /^[vh][cC]ard$/
           result_set << 'h-card'
 
-        #these aren't actually specified by the backcompat faq, but probably should parse them
+        # these aren't actually specified by the backcompat faq, but probably should parse them
         when  /^h[fF]eed$/
           result_set << 'h-feed'
         when  /^h[nN]ews$/
           result_set << 'h-news'
-
         else
-          if @fmt_classes.include? 'h-entry' and html_class == 'author'
+          if @fmt_classes.include?('h-entry') && html_class == 'author'
             result_set << 'h-card'
           end
-          if @fmt_classes.include? 'h-product' and html_class == 'review'
+
+          if @fmt_classes.include?('h-product') && html_class == 'review'
             result_set << 'h-review'
           end
-          if @fmt_classes.include? 'h-recipe' and html_class == 'author'
+
+          if @fmt_classes.include?('h-recipe') && html_class == 'author'
             result_set << 'h-card'
           end
-          if @fmt_classes.include? 'h-resume' and html_class == 'contact'
+
+          if @fmt_classes.include?('h-resume') && html_class == 'contact'
             result_set << 'h-card'
           end
-          if @fmt_classes.include? 'h-resume' and html_class == 'education'
+
+          if @fmt_classes.include?('h-resume') && html_class == 'education'
             result_set << 'h-event'
           end
-          if @fmt_classes.include? 'h-resume' and html_class == 'experience'
+
+          if @fmt_classes.include?('h-resume') && html_class == 'experience'
             result_set << 'h-event'
           end
-          if @fmt_classes.include? 'h-resume' and html_class == 'affiliation'
+
+          if @fmt_classes.include?('h-resume') && html_class == 'affiliation'
             result_set << 'h-card'
           end
-          if @fmt_classes.include? 'h-review' and html_class == 'reviewer'
+
+          if @fmt_classes.include?('h-review') && html_class == 'reviewer'
             result_set << 'h-card'
           end
-          if @fmt_classes.include? 'h-review' and html_class == 'item'
-            if not html_classes.include? 'vcard' and not html_classes.include? 'vevent' and not html_classes.include? 'hproduct'
+
+          if @fmt_classes.include?('h-review') && html_class == 'item'
+            if !html_classes.include?('vcard') && !html_classes.include?('vevent') && !html_classes.include?('hproduct')
               result_set << 'h-item'
             end
           end
-          if @fmt_classes.include? 'h-review-aggregate' and html_class == 'item'
-            if not html_classes.include? 'vcard' and not html_classes.include? 'vevent' and not html_classes.include? 'hproduct'
+
+          if @fmt_classes.include?('h-review-aggregate') && html_class == 'item'
+            if !html_classes.include?('vcard') && !html_classes.include?('vevent') && !html_classes.include?('hproduct')
               result_set << 'h-item'
             end
           end
-          if @fmt_classes.include? 'h-review-aggregate' and html_class == 'reviewer'
+
+          if @fmt_classes.include?('h-review-aggregate') && html_class == 'reviewer'
             result_set << 'h-card'
           end
-          if @fmt_classes.include? 'h-entry' and html_class == 'location'
+
+          if @fmt_classes.include?('h-entry') && html_class == 'location'
             result_set << 'h-adr'
             result_set << 'h-card'
           end
-          if @fmt_classes.include? 'h-feed' and html_class == 'author'
+
+          if @fmt_classes.include?('h-feed') && html_class == 'author'
             result_set << 'h-card'
           end
         end
-
       end
+
       result_set.uniq
     end
 
@@ -134,195 +140,199 @@ module Microformats
       result_set = []
       rels = element.attribute('rel').to_s.split
 
-      if @fmt_classes.include? 'h-entry' and rels.include? 'bookmark'
+      if @fmt_classes.include?('h-entry') && rels.include?('bookmark')
         result_set << 'u-url'
       end
-      if @fmt_classes.include? 'h-entry' and rels.include? 'tag'
+
+      if @fmt_classes.include?('h-entry') && rels.include?('tag')
         result_set << 'p-category'
       end
-      if @fmt_classes.include? 'h-recipe' and rels.include? 'tag'
+
+      if @fmt_classes.include?('h-recipe') && rels.include?('tag')
         result_set << 'p-category'
       end
-      if @fmt_classes.include? 'h-review' and rels.include? 'tag'
+
+      if @fmt_classes.include?('h-review') && rels.include?('tag')
         result_set << 'p-category'
       end
-      if @fmt_classes.include? 'h-feed' and rels.include? 'tag'
+
+      if @fmt_classes.include?('h-feed') && rels.include?('tag')
         result_set << 'p-category'
       end
-      if @fmt_classes.include? 'h-review' and rels.include? 'self' and rels.include? 'bookmark'
+
+      if @fmt_classes.include?('h-review') && rels.include?('self') && rels.include?('bookmark')
         result_set << 'u-url'
       end
-      if @fmt_classes.include? 'h-news' and rels.include? 'principles'
+
+      if @fmt_classes.include?('h-news') && rels.include?('principles')
         result_set << 'u-principles'
       end
 
-      #TODO PROPOSED convert time.entry-date[datetime] to dt-published  see wiki/h-entry
-      #TODO PROPOSED convert rel=author to u-author  see wiki/h-entry
+      # TODO: PROPOSED convert time.entry-date[datetime] to dt-published  see wiki/h-entry
+      # TODO: PROPOSED convert rel=author to u-author  see wiki/h-entry
 
       element.attribute('class').to_s.split.each do |html_class|
-
-        if @fmt_classes.include? 'h-adr'
-          if [ 'post-office-box', 'extended-address', 'street-address', 'locality', 'region', 'postal-code', 'country-name' ].include? html_class
+        if @fmt_classes.include?('h-adr')
+          if %w[post-office-box extended-address street-address locality region postal-code country-name].include?(html_class)
             result_set << 'p-' + html_class
           end
         end
 
-        if @fmt_classes.include? 'h-geo'
-          if [ 'longitude', 'latitude' ].include? html_class
+        if @fmt_classes.include?('h-geo')
+          if %w[longitude latitude].include?(html_class)
             result_set << 'p-' + html_class
           end
         end
 
-        if @fmt_classes.include? 'h-entry'
-          if html_class ==  'entry-title'
+        if @fmt_classes.include?('h-entry')
+          if html_class == 'entry-title'
             result_set << 'p-name'
-          elsif html_class ==  'entry-summary'
+          elsif html_class == 'entry-summary'
             result_set << 'p-summary'
-          elsif html_class ==  'entry-content'
+          elsif html_class == 'entry-content'
             result_set << 'e-content'
-          elsif ['updated', 'published'].include? html_class
+          elsif %w[updated published].include?(html_class)
             result_set << 'dt-' + html_class
-          elsif [ 'category', 'author' ].include? html_class
+          elsif %w[category author].include?(html_class)
             result_set << 'p-' + html_class
           end
         end
 
-        #h-news isn't even listed in backcompat list, adding to follow test suite
-        if @fmt_classes.include? 'h-news'
-          if [ 'source-org', 'entry', 'dateline', 'geo' ].include? html_class
+        # h-news isn't even listed in backcompat list, adding to follow test suite
+        if @fmt_classes.include?('h-news')
+          if %w[source-org entry dateline geo].include?(html_class)
             result_set << 'p-' + html_class
           end
         end
 
-        if @fmt_classes.include? 'h-feed'
-          if html_class ==  'fn'
+        if @fmt_classes.include?('h-feed')
+          if html_class == 'fn'
             result_set << 'p-name'
-          elsif [ 'author', 'summary' ].include? html_class
+          elsif %w[author summary].include?(html_class)
             result_set << 'p-' + html_class
-          elsif ['photo', 'url'].include? html_class
+          elsif %w[photo url].include?(html_class)
             result_set << 'u-' + html_class
           end
         end
 
-        if @fmt_classes.include? 'h-item'
-          if html_class ==  'fn'
+        if @fmt_classes.include?('h-item')
+          if html_class == 'fn'
             result_set << 'p-name'
-          elsif ['photo', 'url'].include? html_class
+          elsif %w[photo url].include?(html_class)
             result_set << 'u-' + html_class
           end
         end
 
-        if @fmt_classes.include? 'h-product'
-          if html_class ==  'fn'
+        if @fmt_classes.include?('h-product')
+          if html_class == 'fn'
             result_set << 'p-name'
-          elsif ['photo', 'url', 'identifier'].include? html_class
+          elsif %w[photo url identifier].include?(html_class)
             result_set << 'u-' + html_class
-          elsif [ 'brand', 'category', 'price', 'review', 'description' ].include? html_class
+          elsif %w[brand category price review description].include?(html_class)
             result_set << 'p-' + html_class
           end
         end
 
-        if @fmt_classes.include? 'h-recipe'
-          if html_class ==  'fn'
+        if @fmt_classes.include?('h-recipe')
+          if html_class == 'fn'
             result_set << 'p-name'
-          elsif html_class ==  'instructions'
+          elsif html_class == 'instructions'
             result_set << 'e-instructions'
-          elsif html_class ==  'duration'
+          elsif html_class == 'duration'
             result_set << 'dt-duration'
-          elsif html_class ==  'photo'
+          elsif html_class == 'photo'
             result_set << 'u-photo'
-          elsif [ 'ingredient', 'category', 'yield', 'summary', 'nutrition', 'author' ].include? html_class
+          elsif %w[ingredient category yield summary nutrition author].include?(html_class)
             result_set << 'p-' + html_class
           end
         end
 
-        if @fmt_classes.include? 'h-resume'
-          if [ 'skill', 'summary', 'contact', 'education', 'experience', 'affiliation' ].include? html_class
+        if @fmt_classes.include?('h-resume')
+          if %w[skill summary contact education experience affiliation].include?(html_class)
             result_set << 'p-' + html_class
           end
         end
 
-        if @fmt_classes.include? 'h-review'
-          if html_class ==  'summary'
+        if @fmt_classes.include?('h-review')
+          if html_class == 'summary'
             result_set << 'p-name'
-          elsif html_class ==  'fn'
+          elsif html_class == 'fn'
             result_set << 'p-name'
-          elsif html_class ==  'reviewer'
+          elsif html_class == 'reviewer'
             result_set << 'p-author'
-          elsif html_class ==  'dtreviewed'
+          elsif html_class == 'dtreviewed'
             result_set << 'dt-published'
-          elsif html_class ==  'description'
+          elsif html_class == 'description'
             result_set << 'e-content'
-          elsif ['photo', 'url', 'identifier'].include? html_class
+          elsif %w[photo url identifier].include?(html_class)
             result_set << 'u-' + html_class
-          elsif [ 'rating', 'best', 'worst', 'item'].include? html_class
+          elsif %w[rating best worst item].include?(html_class)
             result_set << 'p-' + html_class
           end
         end
 
-        if @fmt_classes.include? 'h-review-aggregate'
-          if html_class ==  'summary'
+        if @fmt_classes.include?('h-review-aggregate')
+          if html_class == 'summary'
             result_set << 'p-name'
-          elsif html_class ==  'fn'
+          elsif html_class == 'fn'
             result_set << 'p-name'
-          elsif html_class ==  'reviewer'
+          elsif html_class == 'reviewer'
             result_set << 'p-author'
-          elsif html_class ==  'dtreviewed'
+          elsif html_class == 'dtreviewed'
             result_set << 'dt-published'
-          elsif html_class ==  'description'
+          elsif html_class == 'description'
             result_set << 'e-content'
-          elsif ['photo', 'url', 'identifier'].include? html_class
+          elsif %w[photo url identifier].include?(html_class)
             result_set << 'u-' + html_class
-          elsif [ 'rating', 'best', 'worst', 'item', 'count', 'votes', 'average' ].include? html_class
+          elsif %w[rating best worst item count votes average].include?(html_class)
             result_set << 'p-' + html_class
           end
         end
 
-        if @fmt_classes.include? 'h-event'
-          if html_class ==  'summary'
+        if @fmt_classes.include?('h-event')
+          if html_class == 'summary'
             result_set << 'p-name'
-          elsif html_class ==  'dtstart'
+          elsif html_class == 'dtstart'
             result_set << 'dt-start'
-          elsif html_class ==  'dtend'
+          elsif html_class == 'dtend'
             result_set << 'dt-end'
-          elsif html_class ==  'duration'
+          elsif html_class == 'duration'
             result_set << 'dt-duration'
-          elsif html_class ==  'geo'
+          elsif html_class == 'geo'
             result_set << 'p-location'
-          elsif ['url'].include? html_class
+          elsif %w[url].include?(html_class)
             result_set << 'u-' + html_class
-          elsif [ 'description', 'category', 'location', 'attendee'].include? html_class
+          elsif %w[description category location attendee].include?(html_class)
             result_set << 'p-' + html_class
           end
         end
 
-        if @fmt_classes.include? 'h-card'
-          if html_class ==  'fn'
+        if @fmt_classes.include?('h-card')
+          if html_class == 'fn'
             result_set << 'p-name'
-          elsif html_class ==  'bday'
+          elsif html_class == 'bday'
             result_set << 'dt-bday'
-          elsif html_class ==  'title'
+          elsif html_class == 'title'
             result_set << 'p-job-title'
-          elsif html_class ==  'rev'
+          elsif html_class == 'rev'
             result_set << 'dt-rev'
-          elsif ['email', 'logo', 'photo', 'url', 'uid', 'key'].include? html_class
+          elsif %w[email logo photo url uid key].include?(html_class)
             result_set << 'u-' + html_class
-          elsif [ 'honorific-prefix', 'given-name', 'additional-name', 'family-name', 'honorific-suffix', 'nickname',
-                  'category', 'adr', 'extended-address', 'street-address', 'locality', 'region', 'postal-code', 'country-name',
-                  'label', 'geo', 'latitude', 'longitude', 'tel', 'note', 'org', 'organization-name', 'organization-unit', 'role', 'tz' ].include? html_class
+          elsif %w[honorific-prefix given-name additional-name family-name honorific-suffix nickname
+                   category adr extended-address street-address locality region postal-code country-name
+                   label geo latitude longitude tel note org organization-name organization-unit role tz].include?(html_class)
             result_set << 'p-' + html_class
 
-          #these aren't listed in the wiki, may be removed
-          elsif ['sound'].include? html_class
+          # these aren't listed in the wiki, may be removed
+          elsif %w[sound].include?(html_class)
             result_set << 'u-' + html_class
-          #these aren't listed in the wiki, may be removed
-          elsif ['agent', 'mailer', 'sort-string', 'class'].include? html_class
+          # these aren't listed in the wiki, may be removed
+          elsif %w[agent mailer sort-string class].include?(html_class)
             result_set << 'p-' + html_class
-
           end
         end
-
       end
+
       result_set.uniq
     end
 
@@ -331,13 +341,36 @@ module Microformats
         html_class =~ VALUE_CLASS_REG_EXP
       end
     end
+
     def value_title_classes(element)
       element.attribute('class').to_s.split.select do |html_class|
         html_class =~ VALUE_TITLE_CLASS_REG_EXP
       end
     end
 
-  end
+    def render_and_strip(data)
+      new_doc = Nokogiri::HTML(data)
+      new_doc.xpath('//script|//style').remove
+      new_doc.text.strip
+    end
 
-end
+    def render_text(in_node, base: nil)
+      doc = Nokogiri::HTML(in_node.inner_html)
+
+      doc.xpath('//script|//style').remove
+
+      # cannot use doc.css('img').each as it makes a copy of them, it does not modify the original
+      doc.traverse do |node|
+        next unless node.name == 'img'
+
+        if !node.attribute('alt').nil?
+          node.replace(node.attribute('alt').value.to_s)
+        elsif !node.attribute('src').nil?
+          node.replace(Microformats::AbsoluteUri.new(node.attribute('src').value.to_s, base: @base).absolutize)
+        end
+      end
 
+      doc.text.strip
+    end
+  end
+end