Add tests for oai-pmh endpoints

.gitlab-ci.yml

 image: ruby:3.0
 
 stages:
+  - test
   - build
 
 variables:
@@ -8,11 +9,17 @@ variables:
   
 before_script:
   - gem install bundler jekyll --no-document
+  - bundle install
+
+test:
+  stage: test
+  script:
+    - bundle exec ruby scripts/check_oai_pmh.rb
+  allow_failure: true
 
 build:
   stage: build
   script:
-    - bundle install
     - bundle exec jekyll build
   artifacts:
     paths:

Gemfile

@@ -2,3 +2,11 @@ source "https://rubygems.org"
 
 # add jekyll as a dependency
 gem "jekyll"
+
+# add dependencies for HTTP requests
+gem "httparty"
+
+# add dependencies to work with yaml and xml files
+gem "yaml"
+gem "nokogiri"
+

Gemfile.lock

@@ -13,6 +13,9 @@ GEM
     forwardable-extended (2.6.0)
     google-protobuf (3.23.3-x86_64-linux)
     http_parser.rb (0.8.0)
+    httparty (0.21.0)
+      mini_mime (>= 1.0.0)
+      multi_xml (>= 0.5.2)
     i18n (1.14.1)
       concurrent-ruby (~> 1.0)
     jekyll (4.3.2)
@@ -44,9 +47,14 @@ GEM
       rb-fsevent (~> 0.10, >= 0.10.3)
       rb-inotify (~> 0.9, >= 0.9.10)
     mercenary (0.4.0)
+    mini_mime (1.1.5)
+    multi_xml (0.6.0)
+    nokogiri (1.16.2-x86_64-linux)
+      racc (~> 1.4)
     pathutil (0.16.2)
       forwardable-extended (~> 2.6)
     public_suffix (5.0.1)
+    racc (1.7.3)
     rake (13.0.6)
     rb-fsevent (0.11.2)
     rb-inotify (0.10.1)
@@ -61,12 +69,16 @@ GEM
       unicode-display_width (>= 1.1.1, < 3)
     unicode-display_width (2.4.2)
     webrick (1.8.1)
+    yaml (0.3.0)
 
 PLATFORMS
   x86_64-linux
 
 DEPENDENCIES
+  httparty
   jekyll
+  nokogiri
+  yaml
 
 BUNDLED WITH
    2.4.15

_data/facilities.yml

@@ -249,9 +249,9 @@
   odr:
     link: https://rodare.hzdr.de/
     oai-pmh-endpoint: 
-      link: https://www.hzdr.de/publications/OAI-PMH?verb=Identify
+      link: https://rodare.hzdr.de/oai2d?verb=Identify
       status: Active
-      last-check: 2024-02-07
+      last-check: 2024-02-19
 
 ## ILL ##
 - short-name: ILL

scripts/check_oai_pmh.rb

+# check_oai_pmh.rb
+
+require 'httparty'
+require 'nokogiri'
+require 'yaml'
+
+def check_oai_pmh_endpoint(endpoint_url)
+    begin
+        response = HTTParty.head(endpoint_url)
+        if response.success?
+            response = HTTParty.get(endpoint_url)
+            if !(response.body.nil? || response.body.empty?)
+                xml_response = Nokogiri::XML(response.body)
+                oai_pmh_tag = xml_response.at_xpath('//*[name()="OAI-PMH"]')
+                if oai_pmh_tag
+                    puts "Endpoint is functional. OAI-PMH tag found in response."
+                    return "Active"
+                else
+                    puts "Endpoint does not contain OAI-PMH tag."
+                    return "Error"
+                end
+            else
+                puts "Endpoint returned an empty body."
+                return "Error"
+            end
+        else
+            puts "Endpoint returned HTTP status code #{response.code}."
+            return "Error"
+        end
+    rescue StandardError => e
+        puts "Error occurred while accessing endpoint: #{e.message}"
+        return "Error"
+    end
+end
+
+up_to_date = true
+facilities = YAML.safe_load(File.read('_data/facilities.yml'), permitted_classes: [Date])
+
+facilities.each do |facility|
+    odr = facility['odr']
+    if odr && odr.key?('oai-pmh-endpoint')
+        oai_pmh_endpoint = odr['oai-pmh-endpoint']['link']
+        if oai_pmh_endpoint
+            expected_status = odr['oai-pmh-endpoint']['status']
+            puts "Checking OAI-PMH endpoint for #{facility['short-name']}: #{oai_pmh_endpoint}"
+            endpoint_status = check_oai_pmh_endpoint(oai_pmh_endpoint)
+            status_match = expected_status == endpoint_status
+            puts "Error: Expected status '#{expected_status}' for #{facility['short-name']} OAI-PMH endpoint, but found status '#{endpoint_status}'." unless status_match
+            up_to_date &= status_match
+        end
+    end
+end
+
+exit(1) unless up_to_date