Add workaround for YouTube Shorts links (#17869)

* Add workaround for YouTube Shorts links * Update link_details_extractor_spec.rb
2022-03-25 19:31:35 +01:00 · 2022-03-25 19:31:35 +01:00 · 79d186ffa8
parent d85a5fa593
commit 79d186ffa8
2 changed files with 9 additions and 1 deletions
--- a/app/lib/link_details_extractor.rb
+++ b/app/lib/link_details_extractor.rb
@ -208,7 +208,7 @@ class LinkDetailsExtractor
  end

  def valid_url_or_nil(str, same_origin_only: false)
-    return if str.blank?
+    return if str.blank? || str == 'null'

    url = @original_url + Addressable::URI.parse(str)

--- a/spec/lib/link_details_extractor_spec.rb
+++ b/spec/lib/link_details_extractor_spec.rb
@ -25,6 +25,14 @@ RSpec.describe LinkDetailsExtractor do
        expect(subject.canonical_url).to eq 'https://foo.com/article'
      end
    end
+
+    context 'when canonical URL is set to "null"' do
+      let(:html) { '<!doctype html><link rel="canonical" href="null" />' }
+
+      it 'ignores the canonical URLs' do
+        expect(subject.canonical_url).to eq original_url
+      end
+    end
  end

  context 'when structured data is present' do