github-mirror/discourse
2
0
Fork 0
mirror of https://github.com/discourse/discourse.git synced 2024-11-25 07:51:36 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

FIX: Canonical URLs may be relative (#14825)

Browse Source 
FinalDestination's follow_canonical mode used for embedded topics should work when canonical URLs are relative, as specified in [RFC 6596](https://datatracker.ietf.org/doc/html/rfc6596)
This commit is contained in:
Roman Rizzi 2021-11-05 14:20:14 -03:00 committed by GitHub
parent f444e3e3f7
commit 53abcd825d
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 36 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
lib/final_destination.rb
View File

@ -225,7 +225,7 @@ class FinalDestination
end end
if @follow_canonical if @follow_canonical
next_url = uri(fetch_canonical_url(response.body)) next_url = fetch_canonical_url(response.body)
if next_url.to_s.present? && next_url != @uri if next_url.to_s.present? && next_url != @uri
@follow_canonical = false @follow_canonical = false
@ -481,10 +481,17 @@ class FinalDestination
def fetch_canonical_url(body) def fetch_canonical_url(body)
return if body.blank? return if body.blank?
canonical_link = Nokogiri::HTML5(body).at("link[rel='canonical']")
return if canonical_link.nil? canonical_element = Nokogiri::HTML5(body).at("link[rel='canonical']")
return if canonical_element.nil?
canonical_uri = uri(canonical_element['href'])
return if canonical_uri.blank?
canonical_link['href'] return canonical_uri if canonical_uri.host.present?
parts = [@uri.host, canonical_uri.to_s]
complete_url = canonical_uri.to_s.starts_with?('/') ? parts.join('') : parts.join('/')
complete_url = "#{@uri.scheme}://#{complete_url}" if @uri.scheme
uri(complete_url)
end end
end end

25
spec/components/final_destination_spec.rb
View File

@ -194,6 +194,31 @@ describe FinalDestination do
expect(final.status).to eq(:resolved) expect(final.status).to eq(:resolved)
end end
it 'resolves the canonical link when the URL is relative' do
host = "https://codinghorror.com"
canonical_follow("#{host}/blog", "/blog/canonical")
stub_request(:head, "#{host}/blog/canonical").to_return(doc_response)
final = FinalDestination.new("#{host}/blog", opts.merge(follow_canonical: true))
expect(final.resolve.to_s).to eq("#{host}/blog/canonical")
expect(final.redirected?).to eq(false)
expect(final.status).to eq(:resolved)
end
it 'resolves the canonical link when the URL is relative and does not start with the / symbol' do
host = "https://codinghorror.com"
canonical_follow("#{host}/blog", "blog/canonical")
stub_request(:head, "#{host}/blog/canonical").to_return(doc_response)
final = FinalDestination.new("#{host}/blog", opts.merge(follow_canonical: true))
expect(final.resolve.to_s).to eq("#{host}/blog/canonical")
expect(final.redirected?).to eq(false)
expect(final.status).to eq(:resolved)
end
it "does not follow the canonical link if it's the same as the current URL" do it "does not follow the canonical link if it's the same as the current URL" do
canonical_follow("https://eviltrout.com", "https://eviltrout.com") canonical_follow("https://eviltrout.com", "https://eviltrout.com")