Feature: Add submission metrics to the indexed data

Gemfile.lock

@@ -1,7 +1,7 @@
 GIT
   remote: https://github.com/ontoportal-lirmm/goo.git
-  revision: 8ddd2d719617ad082c6964a9efdac153cdd2b48e
-  branch: master
+  revision: 0e554fce49713ce4d5a742a06c2fb59a547caf47
+  branch: development
   specs:
     goo (0.0.2)
       addressable (~> 2.8)
@@ -41,7 +41,7 @@ GEM
     ast (2.4.2)
     base64 (0.2.0)
     bcrypt (3.1.20)
-    bigdecimal (3.1.6)
+    bigdecimal (3.1.7)
     builder (3.2.4)
     coderay (1.1.3)
     concurrent-ruby (1.2.3)
@@ -114,7 +114,7 @@ GEM
     method_source (1.0.0)
     mime-types (3.5.2)
       mime-types-data (~> 3.2015)
-    mime-types-data (3.2024.0206)
+    mime-types-data (3.2024.0305)
     mini_mime (1.1.5)
     minitest (4.7.5)
     minitest-reporters (0.14.24)
@@ -132,7 +132,7 @@ GEM
       net-protocol
     net-protocol (0.2.2)
       timeout
-    net-smtp (0.4.0.1)
+    net-smtp (0.5.0)
       net-protocol
     netrc (0.11.0)
     oj (2.18.5)
@@ -149,7 +149,7 @@ GEM
     pry (0.14.2)
       coderay (~> 1.1)
       method_source (~> 1.0)
-    public_suffix (5.0.4)
+    public_suffix (5.0.5)
     racc (1.7.3)
     rack (1.6.13)
     rack-test (0.8.3)
@@ -173,7 +173,7 @@ GEM
       rexml (~> 3.2)
     redis (5.1.0)
       redis-client (>= 0.17.0)
-    redis-client (0.20.0)
+    redis-client (0.21.1)
       connection_pool
     regexp_parser (2.9.0)
     request_store (1.6.0)
@@ -186,18 +186,18 @@ GEM
     rexml (3.2.6)
     rsolr (1.1.2)
       builder (>= 2.1.2)
-    rubocop (1.61.0)
+    rubocop (1.62.1)
       json (~> 2.3)
       language_server-protocol (>= 3.17.0)
       parallel (~> 1.10)
       parser (>= 3.3.0.2)
       rainbow (>= 2.2.2, < 4.0)
       regexp_parser (>= 1.8, < 3.0)
       rexml (>= 3.2.5, < 4.0)
-      rubocop-ast (>= 1.30.0, < 2.0)
+      rubocop-ast (>= 1.31.1, < 2.0)
       ruby-progressbar (~> 1.7)
       unicode-display_width (>= 2.4.0, < 3.0)
-    rubocop-ast (1.31.1)
+    rubocop-ast (1.31.2)
       parser (>= 3.3.0.4)
     ruby-progressbar (1.13.0)
     ruby2_keywords (0.0.5)
@@ -224,7 +224,7 @@ GEM
     unicode-display_width (2.5.0)
     uuid (2.3.9)
       macaddr (~> 1.0)
-    webmock (3.20.0)
+    webmock (3.23.0)
       addressable (>= 2.8.0)
       crack (>= 0.3.2)
       hashdiff (>= 0.4.0, < 2.0.0)

lib/ontologies_linked_data/concerns/ontology_submissions/submission_index_all_data.rb

@@ -22,7 +22,7 @@ def self.included(base)
         end
 
         def index_sorted_ids(ids, ontology, conn, logger, commit = true)
-          total_triples = Parallel.map(ids.each_slice(100), in_threads: 10) do |ids_slice|
+          total_triples = Parallel.map(ids.each_slice(1000), in_threads: 10) do |ids_slice|
             index_ids = 0
             triples_count = 0
             documents = {}
@@ -48,7 +48,7 @@ def index_sorted_ids(ids, ontology, conn, logger, commit = true)
 
         def index_all_data(logger, commit = true)
           page = 1
-          size = 1000
+          size = 10_000
           count_ids = 0
           total_time = 0
           total_triples = 0

lib/ontologies_linked_data/models/contact.rb

@@ -8,7 +8,10 @@ class Contact < LinkedData::Models::Base
       embedded true
 
       def embedded_doc
-        "#{self.name} #{self.email}"
+        bring(:name) if bring?(:name)
+        bring(:email) if bring?(:email)
+
+        "#{self.name} | #{self.email}"
       end
 
     end

lib/ontologies_linked_data/models/metric.rb

@@ -53,6 +53,14 @@ def self.metrics_id_generator(m)
         raise ArgumentError, "Metrics id needs to be set"
         #return RDF::URI.new(m.submission.id.to_s + "/metrics")
       end
+
+      def embedded_doc
+        doc = indexable_object
+        doc.delete(:resource_model)
+        doc.delete(:resource_id)
+        doc.delete(:id)
+        doc
+      end
     end
   end
 end

test/models/test_search.rb

@@ -14,7 +14,8 @@ def setup
 
   def test_search_ontology
     ont_count, ont_acronyms, created_ontologies = create_ontologies_and_submissions({
-                                                                                      process_submission: false,
+                                                                                      process_submission: true,
+                                                                                      process_options: {process_rdf: true, extract_metadata: false, run_metrics: true},
                                                                                       acronym: 'BROTEST',
                                                                                       name: 'ontTEST Bla',
                                                                                       file_path: '../../../../test/data/ontology_files/BRO_v3.2.owl',
@@ -23,6 +24,7 @@ def test_search_ontology
                                                                                     })
 
     ontologies = LinkedData::Models::Ontology.search('*:*', { fq: 'resource_model: "ontology"' })['response']['docs']
+
     assert_equal 3, ontologies.size
     ontologies.each do |ont|
       select_ont = created_ontologies.select { |ont_created| ont_created.id.to_s.eql?(ont['id']) }.first
@@ -36,7 +38,6 @@ def test_search_ontology
 
     submissions = LinkedData::Models::Ontology.search('*:*', { fq: 'resource_model: "ontology_submission"' })['response']['docs']
     assert_equal 9, submissions.size
-
     submissions.each do |sub|
       created_sub = LinkedData::Models::OntologySubmission.find(RDF::URI.new(sub['id'])).first&.bring_remaining
       refute_nil created_sub
@@ -55,7 +56,20 @@ def test_search_ontology
       assert_equal sub['openSearchDescription_t'], created_sub.openSearchDescription
       assert_equal sub['endpoint_txt'], created_sub.endpoint
       assert_equal sub['uploadFilePath_t'], created_sub.uploadFilePath
-      assert_equal sub['submissionStatus_txt'], created_sub.submissionStatus.map(&:id)
+      assert_equal sub['submissionStatus_txt'].sort, created_sub.submissionStatus.map{|x| x.id.to_s}.sort
+
+      created_sub.metrics.bring_remaining
+
+      assert_equal sub['metrics_classes_i'], created_sub.metrics.classes
+      assert_equal sub['metrics_individuals_i'], created_sub.metrics.individuals
+      assert_equal sub['metrics_properties_i'], created_sub.metrics.properties
+      assert_equal sub['metrics_maxDepth_i'], created_sub.metrics.maxDepth
+      assert_equal sub['metrics_maxChildCount_i'], created_sub.metrics.maxChildCount
+      assert_equal sub['metrics_averageChildCount_i'], created_sub.metrics.averageChildCount
+      assert_equal sub['metrics_classesWithOneChild_i'], created_sub.metrics.classesWithOneChild
+      assert_equal sub['metrics_classesWithMoreThan25Children_i'], created_sub.metrics.classesWithMoreThan25Children
+      assert_equal sub['metrics_classesWithNoDefinition_i'], created_sub.metrics.classesWithNoDefinition
+
       embed_doc = created_sub.ontology.bring_remaining.embedded_doc
       embed_doc.each do |k, v|
         if v.is_a?(Array)