From 21d1ee106526f129d6b369da94814851cbd236c3 Mon Sep 17 00:00:00 2001
From: Bianca Nenciu <nbianca@users.noreply.github.com>
Date: Wed, 14 Apr 2021 18:09:55 +0300
Subject: [PATCH] FIX: Use Nokogiri and Loofah consistently (#12693)

CookedPostProcessor used Loofah to parse the cooked content of a post
and Nokogiri to parse cooked Oneboxes. Even though Loofah is built on
top of Nokogiri, replacing an element from the cooked post (a Nokogiri
node) with a parsed onebox (a Loofah node) produced a strange result
which included XML namespaces. Removing the mix and using Loofah
to parse Oneboxes fixed the problem.
---
 lib/oneboxer.rb                  |  9 +++------
 spec/components/oneboxer_spec.rb | 11 +++++++++++
 2 files changed, 14 insertions(+), 6 deletions(-)

diff --git a/lib/oneboxer.rb b/lib/oneboxer.rb
index bdca6e35749..4e0b96ef9aa 100644
--- a/lib/oneboxer.rb
+++ b/lib/oneboxer.rb
@@ -112,10 +112,7 @@ module Oneboxer
   end
 
   # Parse URLs out of HTML, returning the document when finished.
-  def self.each_onebox_link(string_or_doc, extra_paths: [])
-    doc = string_or_doc
-    doc = Nokogiri::HTML5::fragment(doc) if doc.is_a?(String)
-
+  def self.each_onebox_link(doc, extra_paths: [])
     onebox_links = doc.css("a.#{ONEBOX_CSS_CLASS}", *extra_paths)
     if onebox_links.present?
       onebox_links.each do |link|
@@ -130,14 +127,14 @@ module Oneboxer
 
   def self.apply(string_or_doc, extra_paths: nil)
     doc = string_or_doc
-    doc = Nokogiri::HTML5::fragment(doc) if doc.is_a?(String)
+    doc = Loofah.fragment(doc) if doc.is_a?(String)
     changed = false
 
     each_onebox_link(doc, extra_paths: extra_paths) do |url, element|
       onebox, _ = yield(url, element)
       next if onebox.blank?
 
-      parsed_onebox = Nokogiri::HTML5::fragment(onebox)
+      parsed_onebox = Loofah.fragment(onebox)
       next if parsed_onebox.children.blank?
 
       changed = true
diff --git a/spec/components/oneboxer_spec.rb b/spec/components/oneboxer_spec.rb
index 1868ec20b6d..b3c27c52e06 100644
--- a/spec/components/oneboxer_spec.rb
+++ b/spec/components/oneboxer_spec.rb
@@ -328,6 +328,17 @@ describe Oneboxer do
         <p>After Onebox</p>
       HTML
     end
+
+    it 'does keeps SVGs valid' do
+      raw = "Onebox\n\nhttps://example.com"
+      cooked = PrettyText.cook(raw)
+      cooked = Oneboxer.apply(Loofah.fragment(cooked)) { '<div><svg><path></path></svg></div>' }
+      doc = Nokogiri::HTML5::fragment(cooked.to_html)
+      expect(doc.to_html).to match_html <<~HTML
+        <p>Onebox</p>
+        <div><svg><path></path></svg></div>
+      HTML
+    end
   end
 
   describe '#force_get_hosts' do