FEATURE: Nokogumbo (#9577)

* FEATURE: Nokogumbo

Use Nokogumbo HTML parser.

app/models/category.rb

@@ -306,7 +306,7 @@ class Category < ActiveRecord::Base
 
     @@cache_text ||= LruRedux::ThreadSafeCache.new(1000)
     @@cache_text.getset(self.description) do
-      text = Nokogiri::HTML.fragment(self.description).text.strip
+      text = Nokogiri::HTML5.fragment(self.description).text.strip
       Rack::Utils.escape_html(text).html_safe
     end
   end

app/models/post.rb

@@ -953,7 +953,7 @@ class Post < ActiveRecord::Base
       /\/uploads\/short-url\/[a-zA-Z0-9]+(\.[a-z0-9]+)?/
     ]
 
-    fragments ||= Nokogiri::HTML::fragment(self.cooked)
+    fragments ||= Nokogiri::HTML5::fragment(self.cooked)
     selectors = fragments.css("a/@href", "img/@src", "source/@src", "track/@src", "video/@poster")
 
     links = selectors.map do |media|

app/models/post_analyzer.rb

@@ -131,7 +131,7 @@ class PostAnalyzer
 
   def cooked_stripped
     @cooked_stripped ||= begin
-      doc = Nokogiri::HTML.fragment(cook(@raw, topic_id: @topic_id))
+      doc = Nokogiri::HTML5.fragment(cook(@raw, topic_id: @topic_id))
       doc.css("pre .mention, aside.quote > .title, aside.quote .mention, aside.quote .mention-group, .onebox, .elided").remove
       doc
     end

app/models/quoted_post.rb

@@ -9,7 +9,7 @@ class QuotedPost < ActiveRecord::Base
   #  we are double parsing this fragment, this may be worth optimising later
   def self.extract_from(post)
 
-    doc = Nokogiri::HTML.fragment(post.cooked)
+    doc = Nokogiri::HTML5.fragment(post.cooked)
 
     uniq = {}
 

app/models/theme_field.rb

@@ -78,7 +78,7 @@ class ThemeField < ActiveRecord::Base
 
     js_compiler = ThemeJavascriptCompiler.new(theme_id, self.theme.name)
 
-    doc = Nokogiri::HTML.fragment(html)
+    doc = Nokogiri::HTML5.fragment(html)
 
     doc.css('script[type="text/x-handlebars"]').each do |node|
       name = node["name"] || node["data-template-name"] || "broken"

app/models/topic_embed.rb

@@ -126,7 +126,7 @@ class TopicEmbed < ActiveRecord::Base
       return
     end
 
-    raw_doc = Nokogiri::HTML(html)
+    raw_doc = Nokogiri::HTML5(html)
     auth_element = raw_doc.at('meta[@name="author"]')
     if auth_element.present?
       response.author = User.where(username_lower: auth_element[:content].strip).first
@@ -142,7 +142,7 @@ class TopicEmbed < ActiveRecord::Base
       title.strip!
     end
     response.title = title
-    doc = Nokogiri::HTML(read_doc.content)
+    doc = Nokogiri::HTML5(read_doc.content)
 
     tags = { 'img' => 'src', 'script' => 'src', 'a' => 'href' }
     doc.search(tags.keys.join(',')).each do |node|
@@ -198,7 +198,7 @@ class TopicEmbed < ActiveRecord::Base
     prefix = "#{uri.scheme}://#{uri.host}"
     prefix += ":#{uri.port}" if uri.port != 80 && uri.port != 443
 
-    fragment = Nokogiri::HTML.fragment("<div>#{contents}</div>")
+    fragment = Nokogiri::HTML5.fragment("<div>#{contents}</div>")
     fragment.css('a').each do |a|
       href = a['href']
       if href.present? && href.start_with?('/')
@@ -220,7 +220,7 @@ class TopicEmbed < ActiveRecord::Base
   end
 
   def self.first_paragraph_from(html)
-    doc = Nokogiri::HTML(html)
+    doc = Nokogiri::HTML5(html)
 
     result = +""
     doc.css('p').each do |p|