From c49164bd1ca81b48155de735730795fd24f690f4 Mon Sep 17 00:00:00 2001
From: Benjamin Armintor <armintor@gmail.com>
Date: Tue, 12 Dec 2023 18:55:56 -0500
Subject: [PATCH] extract a concern for scrubbing resolvers and include in Site
 (DLC-1052)

---
 .../concerns/solr_document/clean_resolver.rb  | 17 ++++++++++++++
 app/models/site.rb                            |  6 +++++
 app/models/solr_document.rb                   | 17 +-------------
 spec/models/site_spec.rb                      | 23 +++++++++++++++++++
 4 files changed, 47 insertions(+), 16 deletions(-)
 create mode 100644 app/models/concerns/solr_document/clean_resolver.rb

diff --git a/app/models/concerns/solr_document/clean_resolver.rb b/app/models/concerns/solr_document/clean_resolver.rb
new file mode 100644
index 00000000..b0e7fd61
--- /dev/null
+++ b/app/models/concerns/solr_document/clean_resolver.rb
@@ -0,0 +1,17 @@
+module SolrDocument::CleanResolver
+  # Scrub permanent links from catalog data to use modern resolver syntax
+  # @param perma_link [String] the original link
+  # @return [String] link with cgi version of resolver replaced with modern version
+  def clean_resolver(link_src)
+    if link_src
+      link_uri = URI(link_src)
+      if link_uri.path == "/cgi-bin/cul/resolve" && link_uri.host == "www.columbia.edu"
+        return "https://resolver.library.columbia.edu/#{link_uri.query}"
+      end
+      if link_uri.host == "library.columbia.edu" && link_uri.path =~ /^\/resolve\/([^\/]+)/
+        return "https://resolver.library.columbia.edu/#{$1}"
+      end
+    end
+    link_src
+  end
+end
diff --git a/app/models/site.rb b/app/models/site.rb
index 551e03a9..c2a0e530 100644
--- a/app/models/site.rb
+++ b/app/models/site.rb
@@ -2,6 +2,7 @@
 class Site < ApplicationRecord
 	include Dcv::Sites::Constants
 	include Blacklight::Configurable
+	include SolrDocument::CleanResolver
 	has_many :scope_filters, as: :scopeable
 	has_many :nav_links, dependent: :destroy, inverse_of: :site
 	has_many :site_pages, dependent: :destroy
@@ -181,6 +182,11 @@ def watermark_url
 		watermark_uploader.store_path.sub(File.join(Rails.root, 'public'), '')
 	end
 
+	# scrub CUL resolvers for format, or pass through
+    def persistent_url
+    	clean_resolver(super)
+    end
+
 	def to_subsite_config
 		config = {
 			'slug' => slug, 'restricted' => (slug =~ /restricted/).present?, 'palette' => palette, 'layout' => layout, 'scope_constraints' => constraints
diff --git a/app/models/solr_document.rb b/app/models/solr_document.rb
index e025433c..9039fe21 100644
--- a/app/models/solr_document.rb
+++ b/app/models/solr_document.rb
@@ -15,6 +15,7 @@ class SolrDocument
   RESOURCE_MODEL = 'GenericResource'
 
   include Blacklight::Solr::Document
+  include SolrDocument::CleanResolver
   include SolrDocument::FieldSemantics
   include SolrDocument::OpenUrlContext
   include SolrDocument::PublicationInfo
@@ -89,22 +90,6 @@ def schema_image_identifier
     end
   end
 
-  # Scrub permanent links from catalog data to use modern resolver syntax
-  # @param perma_link [String] the original link
-  # @return [String] link with cgi version of resolver replaced with modern version
-  def clean_resolver(link_src)
-    if link_src
-      link_uri = URI(link_src)
-      if link_uri.path == "/cgi-bin/cul/resolve" && link_uri.host == "www.columbia.edu"
-        return "https://resolver.library.columbia.edu/#{link_uri.query}"
-      end
-      if link_uri.host == "library.columbia.edu" && link_uri.path =~ /^\/resolve\/([^\/]+)/
-        return "https://resolver.library.columbia.edu/#{$1}"
-      end
-    end
-    link_src
-  end
-
   def slug
     if self[:restriction_ssim].present?
       Array(self[:slug_ssim]).compact.map { |val| "restricted/#{val}"  }.first
diff --git a/spec/models/site_spec.rb b/spec/models/site_spec.rb
index b3e796d9..1b5950d2 100644
--- a/spec/models/site_spec.rb
+++ b/spec/models/site_spec.rb
@@ -249,4 +249,27 @@
 			end
 		end
 	end
+	describe '#persistent_url' do
+		let(:site_slug) { 'persistent_url' }
+		let(:rkey) { 'lweb0138' }
+		let(:cgi_http) { "http://www.columbia.edu/cgi-bin/cul/resolve?#{rkey}" }
+		let(:cgi_https) { "https://www.columbia.edu/cgi-bin/cul/resolve?#{rkey}" }
+		let(:lweb_http) { "https://library.columbia.edu/resolve/#{rkey}" }
+		let(:lweb_https) { "https://library.columbia.edu/resolve/#{rkey}" }
+		let(:current_https) { "https://resolver.library.columbia.edu/#{rkey}" }
+		let(:na_https) { "https://nothing.library.columbia.edu/#{rkey}" }
+		it "cleans resolvers" do
+			site.persistent_url = cgi_http
+			expect(site.persistent_url).to eql(current_https)
+			site.persistent_url = cgi_https
+			expect(site.persistent_url).to eql(current_https)
+			site.persistent_url = lweb_http
+			expect(site.persistent_url).to eql(current_https)
+			site.persistent_url = lweb_https
+			expect(site.persistent_url).to eql(current_https)
+			site.persistent_url = na_https
+			expect(site.persistent_url).to eql(na_https)
+		end
+		
+	end
 end
\ No newline at end of file