From 614dd22095a45ab78d7469b1a034b7277878a639 Mon Sep 17 00:00:00 2001 From: Claire Date: Wed, 11 Jan 2023 22:21:10 +0100 Subject: [PATCH] Fix sanitizer parsing link text as HTML when stripping unsupported links (#22558) --- lib/sanitize_ext/sanitize_config.rb | 2 +- spec/lib/sanitize_config_spec.rb | 4 ++++ 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/lib/sanitize_ext/sanitize_config.rb b/lib/sanitize_ext/sanitize_config.rb index a2e1d9d01b..baf6526626 100644 --- a/lib/sanitize_ext/sanitize_config.rb +++ b/lib/sanitize_ext/sanitize_config.rb @@ -49,7 +49,7 @@ class Sanitize end end - current_node.replace(current_node.text) unless LINK_PROTOCOLS.include?(scheme) + current_node.replace(Nokogiri::XML::Text.new(current_node.text, current_node.document)) unless LINK_PROTOCOLS.include?(scheme) end UNSUPPORTED_ELEMENTS_TRANSFORMER = lambda do |env| diff --git a/spec/lib/sanitize_config_spec.rb b/spec/lib/sanitize_config_spec.rb index 747d81158d..c9543ceb0c 100644 --- a/spec/lib/sanitize_config_spec.rb +++ b/spec/lib/sanitize_config_spec.rb @@ -38,6 +38,10 @@ describe Sanitize::Config do expect(Sanitize.fragment('Test', subject)).to eq 'Test' end + it 'does not re-interpret HTML when removing unsupported links' do + expect(Sanitize.fragment('Test<a href="https://example.com">test</a>', subject)).to eq 'Test<a href="https://example.com">test</a>' + end + it 'keeps a with href' do expect(Sanitize.fragment('Test', subject)).to eq 'Test' end