Fix sanitizer making block level elements unreadable (#10836)

Fix #10834

2 files changed, 41 insertions, 0 deletions

diff --git a/app/lib/sanitize_config.rb b/app/lib/sanitize_config.rb
index 1bba4a5a6..e82a2a33a 100644
--- a/app/lib/sanitize_config.rb
+++ b/app/lib/sanitize_config.rb
@@ -19,6 +19,20 @@ class Sanitize
       node['class'] = class_list.join(' ')
     end
 
+    UNSUPPORTED_ELEMENTS_TRANSFORMER = lambda do |env|
+      return unless %w(h1 h2 h3 h4 h5 h6 blockquote pre ul ol li).include?(env[:node_name])
+
+      case env[:node_name]
+      when 'li'
+        env[:node].traverse do |node|
+          node.add_next_sibling('<br>') if node.next_sibling
+          node.replace(node.children) unless node.text?
+        end
+      else
+        env[:node].name = 'p'
+      end
+    end
+
     MASTODON_STRICT ||= freeze_config(
       elements: %w(p br span a),
 
@@ -40,6 +54,7 @@ class Sanitize
 
       transformers: [
         CLASS_WHITELIST_TRANSFORMER,
+        UNSUPPORTED_ELEMENTS_TRANSFORMER,
       ]
     )
 
diff --git a/spec/lib/sanitize_config_spec.rb b/spec/lib/sanitize_config_spec.rb
new file mode 100644
index 000000000..bb3cf6f0b
--- /dev/null
+++ b/spec/lib/sanitize_config_spec.rb
@@ -0,0 +1,26 @@
+# frozen_string_literal: true
+
+require 'rails_helper'
+require Rails.root.join('app', 'lib', 'sanitize_config.rb')
+
+describe Sanitize::Config do
+  describe '::MASTODON_STRICT' do
+    subject { Sanitize::Config::MASTODON_STRICT }
+
+    it 'converts h1 to p' do
+      expect(Sanitize.fragment('<h1>Foo</h1>', subject)).to eq '<p>Foo</p>'
+    end
+
+    it 'converts ul to p' do
+      expect(Sanitize.fragment('<p>Check out:</p><ul><li>Foo</li><li>Bar</li></ul>', subject)).to eq '<p>Check out:</p><p>Foo<br>Bar</p>'
+    end
+
+    it 'converts p inside ul' do
+      expect(Sanitize.fragment('<ul><li><p>Foo</p><p>Bar</p></li><li>Baz</li></ul>', subject)).to eq '<p>Foo<br>Bar<br>Baz</p>'
+    end
+
+    it 'converts ul inside ul' do
+      expect(Sanitize.fragment('<ul><li>Foo</li><li><ul><li>Bar</li><li>Baz</li></ul></li></ul>', subject)).to eq '<p>Foo<br>Bar<br>Baz</p>'
+    end
+  end
+end