Allow Welsh-language mainstream URLs to be added to document collection groups

The Publishing API's lookup_content_id endpoint does not index translated
(Welsh) base paths, causing a false "must reference a GOV.UK page" error
when users tried to add Welsh mainstream URLs to a document collection.

Fall back to the Content Store when both Publishing API lookups return no
content ID, as the Content Store resolves content items by any translated
base path. If the Content Store also returns a 404, the original error is
still raised.

Adds Services.content_store using GdsApi::ContentStore, and updates tests
to cover the Welsh URL case and to stub the Content Store in the existing
"not found" test.

Author: GDSNewt

app/models/document_collection_non_whitehall_link/govuk_url.rb

@@ -31,7 +31,12 @@ def parsed_url
   end
 
   def content_item
-    @content_item ||= Services.publishing_api.get_content(content_id).to_h
+    content_id
+    @content_item ||= begin
+      Services.publishing_api.get_content(@content_id).to_h
+    rescue GdsApi::HTTPNotFound
+      content_item_from_content_store
+    end
   end
 
   def content_id
@@ -40,15 +45,21 @@ def content_id
     if @content_id.blank?
       toplevel_path_segment = parsed_url.path.split("/").second
       @content_id = Services.publishing_api.lookup_content_id(base_path: "/#{toplevel_path_segment}", with_drafts: true)
+
       if @content_id.blank?
-        raise GdsApi::HTTPNotFound, 404
+        @content_id = content_item_from_content_store["content_id"]
+        raise GdsApi::HTTPNotFound, 404 if @content_id.blank?
       else
-        unless content_item["document_type"] == "guide"
-          raise GdsApi::HTTPNotFound, 404
-        end
+        raise GdsApi::HTTPNotFound, 404 unless content_item["document_type"] == "guide"
       end
     end
 
     @content_id
   end
+
+  def content_item_from_content_store
+    Services.content_store.content_item(parsed_url.path).to_h
+  rescue GdsApi::ContentStore::ItemNotFound
+    raise GdsApi::HTTPNotFound, 404
+  end
 end

lib/services.rb

@@ -1,5 +1,6 @@
 require "gds_api/publishing_api"
 require "gds_api/asset_manager"
+require "gds_api/content_store"
 require "gds_api/search"
 
 module Services
@@ -11,6 +12,10 @@ def self.publishing_api_with_huge_timeout
     @publishing_api_with_huge_timeout ||= publishing_api_client_with_timeout(60)
   end
 
+  def self.content_store
+    @content_store ||= GdsApi::ContentStore.new(Plek.find("content-store"))
+  end
+
   def self.asset_manager
     @asset_manager ||= GdsApi::AssetManager.new(
       Plek.find("asset-manager"),

test/unit/app/models/document_collection_non_whitehall_link/govuk_url_test.rb

@@ -91,10 +91,31 @@ class DocumentCollectionNonWhitehallLink::GovukUrlTest < ActiveSupport::TestCase
       document_collection_group: build(:document_collection_group),
     )
 
+    Services.content_store.stubs(:content_item).raises(GdsApi::ContentStore::ItemNotFound.new(404))
+
     assert_not url.valid?
     assert url.errors.full_messages.include?("Url must reference a GOV.UK page")
   end
 
+  test "should be valid when a Welsh-language GOV.UK URL is used that is not in the Publishing API path reservations" do
+    welsh_content_id = SecureRandom.uuid
+    stub_publishing_api_has_item(
+      content_id: welsh_content_id,
+      title: "Talu cosb hunanasesiad",
+      base_path: "/talu-cosb-hunanasesiad",
+      publishing_app: "publisher",
+    )
+    content_store_response = { "content_id" => welsh_content_id, "title" => "Talu cosb hunanasesiad", "base_path" => "/talu-cosb-hunanasesiad", "publishing_app" => "publisher" }
+    Services.content_store.stubs(:content_item).with("/talu-cosb-hunanasesiad").returns(content_store_response)
+
+    url = DocumentCollectionNonWhitehallLink::GovukUrl.new(
+      url: "https://www.gov.uk/talu-cosb-hunanasesiad",
+      document_collection_group: build(:document_collection_group),
+    )
+
+    assert url.valid?
+  end
+
   test "should be invalid when Publishing API returns a 404" do
     stub_any_publishing_api_call_to_return_not_found