Feat: add extracted once in the metadata extraction

config/schemes/ontology_submission.yml

@@ -290,7 +290,7 @@ copyrightHolder:
   description: [ 
     "SCHEMA: The party holding the legal copyright to the CreativeWork.",
     "DCTERMS: A person or organization owning or managing rights over the resource." ]
-  extractedMetadata: true
+  extractedMetadata: once
 
 ### Description
 
@@ -504,7 +504,7 @@ hasCreator:
     "DOAP: Maintainer of a project, a project leader.",
     "SCHEMA:author: The author of this content or rating.",
     "SCHEMA:creator: The creator/author of this CreativeWork." ]
-  extractedMetadata: true
+  extractedMetadata: once
   metadataMappings: [ "omv:hasCreator", "dc:creator", "dcterms:creator", "foaf:maker", "prov:wasAttributedTo", "doap:maintainer", "pav:authoredBy", "pav:createdBy", "schema:author", "schema:creator" ]
 
 #Contributor
@@ -518,7 +518,7 @@ hasContributor:
     "OMV: Contributors to the creation of the ontology.",
     "PAV: The resource was contributed to by the given agent.",
     "DOAP: Project contributor" ]
-  extractedMetadata: true
+  extractedMetadata: once
   metadataMappings: [ "omv:hasContributor", "dc:contributor", "dcterms:contributor", "doap:helper", "schema:contributor", "pav:contributedBy" ]
 
 #Curator
@@ -529,7 +529,7 @@ curatedBy:
   description: [ 
     "PAV: Specifies an agent specialist responsible for shaping the expression in an appropriate format. Often the primary agent responsible for ensuring the quality of the representation.", 
     "MOD: An ontology that is evaluated by an agent." ]
-  extractedMetadata: true
+  extractedMetadata: once
   metadataMappings: [ "mod:evaluatedBy", "pav:curatedBy" ]
 
 #Translator
@@ -539,7 +539,7 @@ translator:
   helpText: "Organization or person who adapts a creative work to different languages."
   description: [ 
     "SCHEMA: Organization or person who adapts a creative work to different languages, regional differences and technical requirements of a target market, or that translates during some event." ]
-  extractedMetadata: true
+  extractedMetadata: once
   metadataMappings: [ "schema:translator" ]
 
 #Publisher
@@ -551,7 +551,7 @@ publisher:
     "DCTERMS: An entity responsible for making the resource available.",
     "SCHEMA: The publisher of creative work.",
     "ADMS: The name of the agency that issued the identifier." ]
-  extractedMetadata: true
+  extractedMetadata: once
   metadataMappings: [ "dc:publisher", "dcterms:publisher", "schema:publisher", "adms:schemaAgency" ]
 
 #Funded or sponsored by
@@ -563,7 +563,7 @@ fundedBy:
     "MOD: An ontology that is sponsored by and developed under a project.",
     "FOAF: An organization funding a project or person.",
     "SCHEMA: The organization on whose behalf the creator was working." ]
-  extractedMetadata: true
+  extractedMetadata: once
   metadataMappings: [ "foaf:fundedBy", "mod:sponsoredBy", "schema:sourceOrganization" ]
 
 #Endorsed by
@@ -574,7 +574,7 @@ endorsedBy:
   description: [ 
     "MOD: An ontology endorsed by an agent.",
     "OMV: The parties that have expressed support or approval to this ontology." ]
-  extractedMetadata: true
+  extractedMetadata: once
   metadataMappings: [ "omv:endorsedBy", "mod:endorsedBy" ]
 
 ### Community

lib/ontologies_linked_data/services/submission_process/operations/submission_extract_metadata.rb

@@ -72,6 +72,18 @@ def extract_ontology_metadata(logger, user_params)
 
           next unless attr_settings[:extractedMetadata] && attr_not_excluded
 
+          extracted_once = attr_settings[:extractedMetadata] == "once"
+          if extracted_once
+            # if the attribute should be extracted only once, check if it has already been set in the before last submission
+            values_before = check_value_from_before_latest_submission(attr, logger)
+
+            if values_before.present?
+              @submission.send("#{attr}=", values_before)
+              next
+            end
+          end
+
+
           # a boolean to check if a value that should be single have already been extracted
           single_extracted = false
           type = enforce?(attr, :list) ? :list : :string
@@ -283,6 +295,21 @@ def find_or_create_agent(attr, old_val, logger)
         end
         agent
       end
+
+      def check_value_from_before_latest_submission(attr, logger)
+        @ontology ||= @submission.ontology
+        @ontology.bring(submissions: [:submissionId])
+
+        @submissions ||= @ontology.submissions.to_a.sort_by { |s| -s.submissionId }
+        return nil if @submissions.size < 2
+
+        before_last_sub = @submissions[1]
+        before_last_sub.bring(attr)
+
+        before_last_sub.public_send(attr.to_s)
+      end
+
+
     end
   end
 end

rakelib/docker_based_test.rake

@@ -68,7 +68,7 @@ namespace :test do
       ENV["COMPOSE_PROFILES"]="vo"
       Rake::Task["test:docker:up"].invoke
       #
-      unless system("curl -sf  http://localhost:8890/sparql  || exit 1")
+      unless system("curl -sf -o /dev/null http://localhost:8890/sparql  || exit 1")
         printf("waiting for Virtuoso container to initialize")
         sec = 0
         until system("curl -sf  http://localhost:8890/sparql || exit 1") do
@@ -99,7 +99,7 @@ namespace :test do
       #system("docker compose cp ./test/data/graphdb-repo-config.ttl graphdb:/opt/graphdb/dist/configs/templates/graphdb-repo-config.ttl")
       #system("docker compose cp ./test/data/graphdb-test-load.nt graphdb:/opt/graphdb/dist/configs/templates/graphdb-test-load.nt")
       #system('docker compose exec graphdb sh -c "importrdf load -f -c /opt/graphdb/dist/configs/templates/graphdb-repo-config.ttl -m parallel /opt/graphdb/dist/configs/templates/graphdb-test-load.nt ;"')
-      unless system("curl -sf http://localhost:7200/repositories || exit 1")
+      unless system("curl -o /dev/null -sf http://localhost:7200/repositories || exit 1")
         printf("waiting for Graphdb container to initialize")
         sec = 0
         until system("curl -sf http://localhost:7200/repositories || exit 1") do

test/models/test_ontology_submission.rb

@@ -1161,6 +1161,51 @@ def test_submission_extract_metadata
     end
   end
 
+  def test_submission_extract_metadata_once
+    # This test is only testing one attributefields which is :hasCreator
+    # TO-DO: make generic to all the fields that has extracted once
+    submission_parse("AGROOE", "AGROOE Test extract metadata ontology",
+                      "./test/data/ontology_files/agrooeMappings-05-05-2016.owl", 1,
+                      process_rdf: true, extract_metadata: true, generate_missing_labels: false, delete: true)
+    ont = LinkedData::Models::Ontology.find("AGROOE").first
+    sub1 = ont.latest_submission
+    refute_nil sub1
+
+    sub1.bring_remaining
+    hasCreator_old_values = ["Alfred DC", "Clement Jonquet", "Gaston Dcterms", "Huguette Doap", "Mirabelle Prov", "Paul Foaf", "Vincent Emonet"]
+    assert_equal hasCreator_old_values.sort, sub1.hasCreator.map { |x| x.bring_remaining.name }.sort    
+    sub1.save
+
+    agents_number = LinkedData::Models::Agent.where.count
+
+    # modifing the names of the agents (the curation phase)
+    hasCreator_old_values.each do |agent_name|
+      agent = LinkedData::Models::Agent.where(name: agent_name).first
+      refute_nil agent
+      agent.bring_remaining
+      agent.name = "#{agent_name}#{rand(1000..9999)}"
+      agent.save
+    end
+
+    assert agents_number, LinkedData::Models::Agent.where.count
+
+    # Create new submission
+    submission_parse("AGROOE", "AGROOE Test extract metadata ontology", "./test/data/ontology_files/agrooeMappings-05-05-2016.owl", 2, 
+                      process_rdf: true, extract_metadata: true, generate_missing_labels: false, delete: false)
+    ont = LinkedData::Models::Ontology.find("AGROOE").first
+    sub2 = ont.latest_submission
+    refute_nil sub2
+
+    sub2.bring_remaining
+    sub2.save
+
+    # Check if there is duplications
+    assert agents_number, LinkedData::Models::Agent.where.count
+    all_sub = LinkedData::Models::Ontology.find("AGROOE").first.bring(:submissions => [:hasCreator => [:name]]).submissions
+    sub0 = all_sub[0]
+    sub1 = all_sub[1]
+    sub0.hasCreator.map { |x| x.name }.sort == sub1.hasCreator.map { |x| x.name }.sort
+  end
 
   def test_submission_delete_remove_files
     #This one has resources wih accents.