bolognese 1.8.17 → 1.9.4

data/lib/bolognese/metadata_utils.rb

@@ -108,12 +108,12 @@ module Bolognese
     end
 
     def reverse
-      { "citation" => Array.wrap(related_identifiers).select { |ri| ri["relationType"] == "IsReferencedBy" }.map do |r| 
+      { "citation" => Array.wrap(related_identifiers).select { |ri| ri["relationType"] == "IsReferencedBy" }.map do |r|
         { "@id" => normalize_doi(r["relatedIdentifier"]),
           "@type" => r["resourceTypeGeneral"] || "ScholarlyArticle",
           "identifier" => r["relatedIdentifierType"] == "DOI" ? nil : to_identifier(r) }.compact
         end.unwrap,
-        "isBasedOn" => Array.wrap(related_identifiers).select { |ri| ri["relationType"] == "IsSupplementTo" }.map do |r| 
+        "isBasedOn" => Array.wrap(related_identifiers).select { |ri| ri["relationType"] == "IsSupplementTo" }.map do |r|
           { "@id" => normalize_doi(r["relatedIdentifier"]),
             "@type" => r["resourceTypeGeneral"] || "ScholarlyArticle",
             "identifier" => r["relatedIdentifierType"] == "DOI" ? nil : to_identifier(r) }.compact
@@ -193,6 +193,7 @@ module Bolognese
         "geo_locations" => geo_locations,
         "funding_references" => funding_references,
         "related_identifiers" => related_identifiers,
+        "related_items" => related_items,
         "schema_version" => schema_version,
         "provider_id" => provider_id,
         "client_id" => client_id,

data/lib/bolognese/readers/bibtex_reader.rb

@@ -31,12 +31,13 @@ module Bolognese
 
         meta = string.present? ? BibTeX.parse(string).first : OpenStruct.new
 
-        schema_org = BIB_TO_SO_TRANSLATIONS[meta.try(:type).to_s] || "ScholarlyArticle"
+        bibtex_type = meta.try(:type).to_s
+        schema_org = BIB_TO_SO_TRANSLATIONS[bibtex_type] || "ScholarlyArticle"
         types = {
-          "resourceTypeGeneral" => Metadata::SO_TO_DC_TRANSLATIONS[schema_org],
+          "resourceTypeGeneral" => Metadata::BIB_TO_DC_TRANSLATIONS[bibtex_type],
           "resourceType" => Bolognese::Utils::BIB_TO_CR_TRANSLATIONS[meta.try(:type).to_s] || meta.try(:type).to_s,
           "schemaOrg" => schema_org,
-          "bibtex" => meta.type.to_s,
+          "bibtex" => bibtex_type,
           "citeproc" => BIB_TO_CP_TRANSLATIONS[meta.try(:type).to_s] || "misc",
           "ris" => BIB_TO_RIS_TRANSLATIONS[meta.try(:type).to_s] || "GEN"
         }.compact
@@ -60,7 +61,7 @@ module Bolognese
         container = if meta.try(:journal).present?
                       first_page = meta.try(:pages).present? ? meta.try(:pages).split("-").map(&:strip)[0] : nil
                       last_page = meta.try(:pages).present? ? meta.try(:pages).split("-").map(&:strip)[1] : nil
-                      
+
                       { "type" => "Journal",
                         "title" => meta.journal.to_s,
                         "identifier" => meta.try(:issn).to_s.presence,

data/lib/bolognese/readers/citeproc_reader.rb

@@ -38,7 +38,7 @@ module Bolognese
         citeproc_type = meta.fetch("type", nil)
         schema_org = CP_TO_SO_TRANSLATIONS[citeproc_type] || "CreativeWork"
         types = {
-          "resourceTypeGeneral" => Bolognese::Utils::SO_TO_DC_TRANSLATIONS[schema_org],
+          "resourceTypeGeneral" => Bolognese::Utils::CP_TO_DC_TRANSLATIONS[citeproc_type],
           "reourceType" => meta.fetch("additionalType", nil),
           "schemaOrg" => schema_org,
           "citeproc" => citeproc_type,

data/lib/bolognese/readers/crossref_reader.rb

@@ -93,7 +93,7 @@ module Bolognese
         resource_type = (resource_type || model).to_s.underscore.camelcase.presence
         schema_org = Bolognese::Utils::CR_TO_SO_TRANSLATIONS[resource_type] || "ScholarlyArticle"
         types = {
-          "resourceTypeGeneral" => Bolognese::Utils::SO_TO_DC_TRANSLATIONS[schema_org],
+          "resourceTypeGeneral" => Bolognese::Utils::CR_TO_DC_TRANSLATIONS[resource_type],
           "resourceType" => resource_type,
           "schemaOrg" => schema_org,
           "citeproc" => Bolognese::Utils::CR_TO_CP_TRANSLATIONS[resource_type] || "article-journal",
@@ -133,10 +133,10 @@ module Bolognese
         # check that date is valid iso8601 date
         date_published = nil unless Date.edtf(date_published.to_h["date"]).present?
         date_updated = nil unless Date.edtf(date_updated.to_h["date"]).present?
-        
+
         dates = [date_published, date_updated].compact
         publication_year = date_published.to_h.fetch("date", "")[0..3].presence
-        
+
         state = meta.present? || read_options.present? ? "findable" : "not_found"
 
         related_identifiers = Array.wrap(crossref_is_part_of(journal_metadata)) + Array.wrap(crossref_references(bibliographic_metadata))

data/lib/bolognese/readers/datacite_reader.rb

@@ -92,15 +92,7 @@ module Bolognese
           "ris" => Bolognese::Utils::CR_TO_RIS_TRANSLATIONS[resource_type.to_s.underscore.camelcase] || Bolognese::Utils::DC_TO_RIS_TRANSLATIONS[resource_type_general.to_s.dasherize] || "GEN"
         }.compact
 
-        titles = Array.wrap(meta.dig("titles", "title")).map do |r|
-          if r.blank?
-            nil
-          elsif r.is_a?(String)
-            { "title" => sanitize(r) }
-          else
-            { "title" => sanitize(r["__content__"]), "titleType" => r["titleType"], "lang" => r["lang"] }.compact
-          end
-        end.compact
+        titles = get_titles(meta)
 
         descriptions = Array.wrap(meta.dig("descriptions", "description")).map do |r|
           if r.blank?
@@ -120,7 +112,7 @@ module Bolognese
             hsh_to_spdx(r)
           end
         end.compact
-        
+
         subjects = Array.wrap(meta.dig("subjects", "subject")).reduce([]) do |sum, subject|
           if subject.is_a?(String)
             sum += name_to_fos(subject)
@@ -194,10 +186,59 @@ module Bolognese
             "schemeType" => ri["schemeType"]
           }.compact
         end
+
+        related_items = Array.wrap(meta.dig("relatedItems", "relatedItem")).map do |ri|
+          rii = ri["relatedItemIdentifier"]
+          if rii["relatedItemIdentifierType"] == "DOI"
+            rid = validate_doi(rii["__content__"].to_s.downcase)
+          else
+            rid = rii["__content__"]
+          end
+
+          relatedItemIdentifier = {
+            "relatedItemIdentifier" => rid,
+            "relatedItemIdentifierType" => rii["relatedItemIdentifierType"],
+            "relatedMetadataScheme" => rii["relatedMetadataScheme"],
+            "schemeURI" => rii["schemeURI"],
+            "schemeType" => rii["schemeType"]
+          }.compact
+
+          {
+            "relationType" => ri["relationType"],
+            "relatedItemType" => ri["relatedItemType"],
+            "relatedItemIdentifier" => relatedItemIdentifier,
+            "creators" => get_authors(Array.wrap(ri.dig("creators", "creator"))),
+            "titles" => get_titles(ri),
+            "publicationYear" => ri["publicationYear"],
+            "volume" => ri["volume"],
+            "issue" => ri["issue"],
+            "number" => ri.dig("number", "__content__"),
+            "numberType" => ri.dig("number", "numberType"),
+            "firstPage" => ri["firstPage"],
+            "lastPage" => ri["lastPage"],
+            "publisher" => ri["publisher"],
+            "edition" => ri["edition"],
+            "contributors" => get_authors(Array.wrap(ri.dig("contributors", "contributor"))),
+          }.compact
+        end
+
         geo_locations = Array.wrap(meta.dig("geoLocations", "geoLocation")).map do |gl|
           if !gl.is_a?(Hash) || gl["geoLocationPoint"].is_a?(String) || gl["geoLocationBox"].is_a?(String) || gl["geoLocationPolygon"].is_a?(String)
             nil
           else
+
+            # Handle scenario where multiple geoLocationPolygons are allowed within a single geoLocation
+            # we want to return an array if it's already an array (i.e. multiple geoLocationPolygons)
+            # vs if it's singular just return the object
+            # This is for backwards compatability to allow both scenarios.
+            if gl.dig("geoLocationPolygon").kind_of?(Array)
+              geoLocationPolygon = gl.dig("geoLocationPolygon").map do |glp|
+                Array.wrap(glp.dig("polygonPoint")).map { |glpp| { "polygonPoint" => glpp } }.compact.presence
+              end.compact.presence
+            else
+              geoLocationPolygon = Array.wrap(gl.dig("geoLocationPolygon", "polygonPoint")).map { |glp| { "polygonPoint" => glp } }.compact.presence
+            end
+
             {
               "geoLocationPoint" => {
                 "pointLatitude" => gl.dig("geoLocationPoint", "pointLatitude"),
@@ -209,7 +250,7 @@ module Bolognese
                 "southBoundLatitude" => gl.dig("geoLocationBox", "southBoundLatitude"),
                 "northBoundLatitude" => gl.dig("geoLocationBox", "northBoundLatitude")
               }.compact.presence,
-              "geoLocationPolygon" => Array.wrap(gl.dig("geoLocationPolygon", "polygonPoint")).map { |glp| { "polygonPoint" => glp } }.compact.presence,
+              "geoLocationPolygon" => geoLocationPolygon,
               "geoLocationPlace" => parse_attributes(gl["geoLocationPlace"], first: true).to_s.strip.presence
             }.compact
           end
@@ -238,6 +279,7 @@ module Bolognese
           "language" => parse_attributes(meta.fetch("language", nil), first: true).to_s.strip.presence,
           "geo_locations" => geo_locations,
           "related_identifiers" => related_identifiers,
+          "related_items" => related_items,
           "formats" => formats,
           "sizes" => sizes,
           "schema_version" => schema_version,
@@ -266,6 +308,21 @@ module Bolognese
           {}
         end
       end
+
+      def get_titles(meta)
+        titles = Array.wrap(meta.dig("titles", "title")).map do |r|
+          if r.blank?
+            nil
+          elsif r.is_a?(String)
+            { "title" => sanitize(r) }
+          else
+            { "title" => sanitize(r["__content__"]), "titleType" => r["titleType"], "lang" => r["lang"] }.compact
+          end
+        end.compact
+
+        titles
+      end
+
     end
   end
 end

data/lib/bolognese/readers/ris_reader.rb

@@ -41,14 +41,14 @@ module Bolognese
         ris_type = meta.fetch("TY", nil) || "GEN"
         schema_org = RIS_TO_SO_TRANSLATIONS[ris_type] || "CreativeWork"
         types = {
-          "resourceTypeGeneral" => Metadata::SO_TO_DC_TRANSLATIONS[schema_org],
+          "resourceTypeGeneral" => Metadata::RIS_TO_DC_TRANSLATIONS[ris_type],
           "schemaOrg" => schema_org,
           "citeproc" => RIS_TO_CP_TRANSLATIONS[schema_org] || "misc",
           "ris" => ris_type
         }.compact
 
         id = normalize_doi(options[:doi] || meta.fetch("DO", nil))
-        
+
         author = Array.wrap(meta.fetch("AU", nil)).map { |a| { "creatorName" => a } }
         date_parts = meta.fetch("PY", nil).to_s.split("/")
         created_date_parts = meta.fetch("Y1", nil).to_s.split("/")
@@ -67,7 +67,7 @@ module Bolognese
         end
         container = if meta.fetch("T2", nil).present?
           { "type" => "Journal",
-            "title" => meta.fetch("T2", nil), 
+            "title" => meta.fetch("T2", nil),
             "identifier" => meta.fetch("SN", nil),
             "volume" => meta.fetch("VL", nil),
             "issue" => meta.fetch("IS", nil),

data/lib/bolognese/utils.rb

@@ -45,16 +45,30 @@ module Bolognese
 
     DC_TO_SO_TRANSLATIONS = {
       "Audiovisual" => "MediaObject",
+      "Book" => "Book",
+      "BookChapter" => "Chapter",
       "Collection" => "Collection",
+      "ComputationalNotebook" => "SoftwareSourceCode",
+      "ConferencePaper" => "Article",
+      "ConferenceProceeding" => "Periodical",
+      "DataPaper" => "Article",
       "Dataset" => "Dataset",
+      "Dissertation" => "Thesis",
       "Event" => "Event",
       "Image" => "ImageObject",
       "InteractiveResource" => nil,
+      "Journal" => "Periodical",
+      "JournalArticle" => "ScholarlyArticle",
       "Model" => nil,
+      "OutputManagementPlan" => nil,
+      "PeerReview" => "Review",
       "PhysicalObject" => nil,
+      "Preprint" => nil,
+      "Report" => "Report",
       "Service" => "Service",
       "Software" => "SoftwareSourceCode",
       "Sound" => "AudioObject",
+      "Standard" => nil,
       "Text" => "ScholarlyArticle",
       "Workflow" => nil,
       "Other" => "CreativeWork",
@@ -65,15 +79,29 @@ module Bolognese
 
     DC_TO_CP_TRANSLATIONS = {
       "Audiovisual" => "motion_picture",
+      "Book" => "book",
+      "BookChapter" => "chapter",
       "Collection" => nil,
+      "ComputationalNotebook" => nil,
+      "ConferencePaper" => "paper_conference",
+      "ConferenceProceeding" => nil,
+      "DataPaper" => "report",
       "Dataset" => "dataset",
+      "Dissertation" => nil,
       "Event" => nil,
       "Image" => "graphic",
       "InteractiveResource" => nil,
+      "Journal" => nil,
+      "JournalArticle" => "article-journal",
       "Model" => nil,
+      "OutputManagementPlan" => nil,
+      "PeerReview" => "review",
       "PhysicalObject" => nil,
+      "Preprint" => nil,
+      "Report" => "report",
       "Service" => nil,
       "Sound" => "song",
+      "Standard" => nil,
       "Text" => "report",
       "Workflow" => nil,
       "Other" => nil
@@ -202,13 +230,44 @@ module Bolognese
       "BookSet" => "book"
     }
 
+    CR_TO_DC_TRANSLATIONS = {
+      "Proceedings" => nil,
+      "ReferenceBook" => nil,
+      "JournalIssue" => "Text",
+      "ProceedingsArticle" => "ConferencePaper",
+      "Other" => "Other",
+      "Dissertation" => "Dissertation",
+      "Dataset" => "Dataset",
+      "EditedBook" => "Book",
+      "JournalArticle" => "JournalArticle",
+      "Journal" => "Journal",
+      "Report" => "Report",
+      "BookSeries" => nil,
+      "ReportSeries" => nil,
+      "BookTrack" => nil,
+      "Standard" => "Standard",
+      "BookSection" => "BookChapter",
+      "BookPart" => nil,
+      "Book" => "Book",
+      "BookChapter" => "BookChapter",
+      "SaComponent" => "Text",
+      "StandardSeries" => "Standard",
+      "Monograph" => "book",
+      "Component" => nil,
+      "ReferenceEntry" => nil,
+      "JournalVolume" => nil,
+      "BookSet" => nil,
+      "PostedContent" => "JournalArticle",
+      "PeerReview" => "PeerReview"
+    }
+
     SO_TO_DC_TRANSLATIONS = {
       "Article" => "Text",
       "AudioObject" => "Sound",
       "Blog" => "Text",
       "BlogPosting" => "Text",
-      "Book" => "Text",
-      "Chapter" => "Text",
+      "Book" => "Book",
+      "Chapter" => "BookChapter",
       "Collection" => "Collection",
       "DataCatalog" => "Dataset",
       "Dataset" => "Dataset",
@@ -216,11 +275,11 @@ module Bolognese
       "ImageObject" => "Image",
       "Movie" => "Audiovisual",
       "PublicationIssue" => "Text",
-      "Report" => "Text",
+      "Report" => "Report",
       "ScholarlyArticle" => "Text",
       "Thesis" => "Text",
       "Service" => "Service",
-      "Review" => "Text",
+      "Review" => "PeerReview",
       "SoftwareSourceCode" => "Software",
       "VideoObject" => "Audiovisual",
       "WebPage" => "Text",
@@ -323,21 +382,73 @@ module Bolognese
 
     DC_TO_RIS_TRANSLATIONS = {
       "Audiovisual" => "MPCT",
+      "Book" => "BOOK",
+      "BookChapter" => "CHAP",
       "Collection" => nil,
+      "ComputationalNotebook" => "COMP",
+      "ConferencePaper" => nil,
+      "ConferenceProceeding" => nil,
+      "DataPaper" => nil,
       "Dataset" => "DATA",
+      "Dissertation" => "THES",
       "Event" => nil,
       "Image" => "FIGURE",
       "InteractiveResource" => nil,
+      "Journal" => nil,
+      "JournalArticle" => "JOUR",
       "Model" => nil,
+      "OutputManagementPlan" => nil,
+      "PeerReview" => nil,
       "PhysicalObject" => nil,
+      "Preprint" => nil,
+      "Report" => "RRPT",
       "Service" => nil,
       "Software" => "COMP",
       "Sound" => "SOUND",
+      "Standard" => nil,
       "Text" => "RPRT",
       "Workflow" => nil,
       "Other" => nil
     }
 
+    RIS_TO_DC_TRANSLATIONS = {
+      "BLOG" => "Text",
+      "GEN" => "Text",
+      "CTLG" => "Collection",
+      "DATA" => "Dataset",
+      "FIGURE" => "Image",
+      "THES" => "Dissertation",
+      "MPCT" => "Audiovisual",
+      "JOUR" => "JournalArticle",
+      "COMP" => "Software",
+      "VIDEO" => "Audiovisual",
+      "ELEC" => "Text"
+    }
+
+    BIB_TO_DC_TRANSLATIONS = {
+      "article" => "JournalArticle",
+      "book" => "Book",
+      "inbook" => "BookChapter",
+      "inproceedings" => nil,
+      "manual" => nil,
+      "misc" => "Other",
+      "phdthesis" => "Dissertation",
+      "proceedings" => "ConferenceProceeding",
+      "techreport" => "Report",
+      "unpublished" => nil
+    }
+
+    CP_TO_DC_TRANSLATIONS = {
+      "song" => "Audiovisual",
+      "post-weblog" => "Text",
+      "dataset" => "Dataset",
+      "graphic" => "Image",
+      "motion_picture" => "Audiovisual",
+      "article-journal" => "JournalArticle",
+      "broadcast" => "Audiovisual",
+      "webpage" => "Text"
+    }
+
     SO_TO_BIB_TRANSLATIONS = {
       "Article" => "article",
       "AudioObject" => "misc",
@@ -415,7 +526,7 @@ module Bolognese
         "ris"
       elsif options[:ext] == ".xml" && Maremma.from_xml(string).to_h.dig("crossref_result", "query_result", "body", "query", "doi_record", "crossref")
         "crossref"
-      elsif options[:ext] == ".xml" && Nokogiri::XML(string, nil, 'UTF-8', &:noblanks).collect_namespaces.find { |k, v| v.start_with?("http://datacite.org/schema/kernel") }      
+      elsif options[:ext] == ".xml" && Nokogiri::XML(string, nil, 'UTF-8', &:noblanks).collect_namespaces.find { |k, v| v.start_with?("http://datacite.org/schema/kernel") }
         "datacite"
       elsif options[:ext] == ".json" && Maremma.from_json(string).to_h.dig("@context").to_s.start_with?("http://schema.org", "https://schema.org")
         "schema_org"
@@ -423,7 +534,7 @@ module Bolognese
         "codemeta"
       elsif options[:ext] == ".json" && Maremma.from_json(string).to_h.dig("schemaVersion").to_s.start_with?("http://datacite.org/schema/kernel")
         "datacite_json"
-      elsif options[:ext] == ".json" && Maremma.from_json(string).to_h.dig("types")
+      elsif options[:ext] == ".json" && Maremma.from_json(string).to_h.dig("types") && Maremma.from_json(string).to_h.dig("publication_year").present?
         "crosscite"
       elsif options[:ext] == ".json" && Maremma.from_json(string).to_h.dig("issued", "date-parts").present?
         "citeproc"
@@ -433,7 +544,7 @@ module Bolognese
     def find_from_format_by_string(string)
       if Maremma.from_xml(string).to_h.dig("crossref_result", "query_result", "body", "query", "doi_record", "crossref").present?
         "crossref"
-      elsif Nokogiri::XML(string, nil, 'UTF-8', &:noblanks).collect_namespaces.find { |k, v| v.start_with?("http://datacite.org/schema/kernel") }  
+      elsif Nokogiri::XML(string, nil, 'UTF-8', &:noblanks).collect_namespaces.find { |k, v| v.start_with?("http://datacite.org/schema/kernel") }
         "datacite"
       elsif Maremma.from_json(string).to_h.dig("@context").to_s.start_with?("http://schema.org", "https://schema.org")
         "schema_org"
@@ -441,7 +552,7 @@ module Bolognese
         "codemeta"
       elsif Maremma.from_json(string).to_h.dig("schema-version").to_s.start_with?("http://datacite.org/schema/kernel")
         "datacite_json"
-      elsif Maremma.from_json(string).to_h.dig("types").present?
+      elsif Maremma.from_json(string).to_h.dig("types").present? && Maremma.from_json(string).to_h.dig("publication_year").present?
         "crosscite"
       elsif Maremma.from_json(string).to_h.dig("issued", "date-parts").present?
         "citeproc"
@@ -523,7 +634,7 @@ module Bolognese
       uri = Addressable::URI.parse(id)
 
       return nil unless uri && uri.host && %w(http https ftp).include?(uri.scheme)
-  
+
       # optionally turn into https URL
       uri.scheme = "https" if options[:https]
 
@@ -654,8 +765,8 @@ module Bolognese
             affiliation_identifier = a["affiliationIdentifier"]
           end
 
-          { 
-            "@type" => "Organization", 
+          {
+            "@type" => "Organization",
             "@id" => affiliation_identifier,
             "name" => name }.compact
         end.unwrap
@@ -677,8 +788,8 @@ module Bolognese
             affiliation_identifier = a["affiliationIdentifier"]
           end
 
-          { 
-            "@type" => "Organization", 
+          {
+            "@type" => "Organization",
             "@id" => affiliation_identifier,
             "name" => name }.compact
         end.unwrap
@@ -692,7 +803,7 @@ module Bolognese
     def to_schema_org_container(element, options={})
       return nil unless (element.is_a?(Hash) || (element.nil? && options[:container_title].present?))
 
-      { 
+      {
         "@id" => element["identifier"],
         "@type" => (options[:type] == "Dataset") ? "DataCatalog" : "Periodical",
         "name" => element["title"] || options[:container_title] }.compact
@@ -700,7 +811,7 @@ module Bolognese
 
     def to_schema_org_identifiers(element, options={})
       Array.wrap(element).map do |ai|
-        { 
+        {
           "@type" => "PropertyValue",
           "propertyID" => ai["identifierType"],
           "value" => ai["identifier"] }
@@ -710,7 +821,7 @@ module Bolognese
     def to_schema_org_relation(related_identifiers: nil, relation_type: nil)
       return nil unless related_identifiers.present? && relation_type.present?
 
-      relation_type = relation_type == "References" ? ["References", "Cites", "Documents"] : [relation_type] 
+      relation_type = relation_type == "References" ? ["References", "Cites", "Documents"] : [relation_type]
 
       Array.wrap(related_identifiers).select { |ri| relation_type.include?(ri["relationType"]) }.map do |r|
         if r["relatedIdentifierType"] == "ISSN" && r["relationType"] == "IsPartOf"
@@ -741,7 +852,7 @@ module Bolognese
 
       Array.wrap(geo_location).reduce([]) do |sum, gl|
         if gl.fetch("geoLocationPoint", nil)
-          sum << { 
+          sum << {
             "@type" => "Place",
             "geo" => {
               "@type" => "GeoCoordinates",
@@ -752,7 +863,7 @@ module Bolognese
         end
 
         if gl.fetch("geoLocationBox", nil)
-          sum << { 
+          sum << {
             "@type" => "Place",
             "geo" => {
               "@type" => "GeoShape",
@@ -765,19 +876,22 @@ module Bolognese
         end
 
         if gl.fetch("geoLocationPolygon", nil)
-          sum << { 
+          sum << {
             "@type" => "Place",
             "geo" => {
               "@type" => "GeoShape",
               "address" => gl["geoLocationPlace"],
-              "polygon" => Array.wrap(gl.dig("geoLocationPolygon")).map do |glp| 
-                [glp.dig("polygonPoint", "pointLongitude"), glp.dig("polygonPoint", "pointLatitude")].compact
-              end.compact }
+              "polygon" => Array.wrap(gl.dig("geoLocationPolygon")).map do |glp|
+                Array.wrap(glp).map do |glpp|
+                  [glpp.dig("polygonPoint", "pointLongitude"), glpp.dig("polygonPoint", "pointLatitude")].compact
+                end.compact
+              end.compact.presence,
+              }
           }
         end
 
         if gl.fetch("geoLocationPlace", nil) && !gl.fetch("geoLocationPoint", nil) && !gl.fetch("geoLocationBox", nil) && !gl.fetch("geoLocationPolygon", nil)
-          sum << { 
+          sum << {
             "@type" => "Place",
             "geo" => {
               "@type" => "GeoCoordinates",
@@ -992,7 +1106,7 @@ module Bolognese
 
     def get_iso8601_date(iso8601_time)
       return nil if iso8601_time.nil?
-      
+
       iso8601_time[0..9]
     end
 
@@ -1089,7 +1203,7 @@ module Bolognese
       first_page = pages.present? ? pages.split("-").map(&:strip)[0] : nil
       last_page = pages.present? ? pages.split("-").map(&:strip)[1] : nil
 
-      { 
+      {
         "title" => title,
         "volume" => volume,
         "issue" => issue,
@@ -1148,11 +1262,11 @@ module Bolognese
     def name_to_fos(name)
       # first find subject in Fields of Science (OECD)
       fos = JSON.load(File.read(File.expand_path('../../../resources/oecd/fos-mappings.json', __FILE__))).fetch("fosFields")
-      
+
       subject = fos.find { |l| l["fosLabel"] == name || "FOS: " + l["fosLabel"] == name }
 
       if subject
-        return [{ 
+        return [{
           "subject" => sanitize(name) },
         {
           "subject" => "FOS: " + subject["fosLabel"],
@@ -1166,12 +1280,12 @@ module Bolognese
       fores = JSON.load(File.read(File.expand_path('../../../resources/oecd/for-mappings.json', __FILE__)))
       for_fields = fores.fetch("forFields")
       for_disciplines = fores.fetch("forDisciplines")
-      
+
       subject = for_fields.find { |l| l["forLabel"] == name } ||
                 for_disciplines.find { |l| l["forLabel"] == name }
 
       if subject
-        [{ 
+        [{
           "subject" => sanitize(name) },
         {
           "subject" => "FOS: " + subject["fosLabel"],
@@ -1189,16 +1303,17 @@ module Bolognese
       subject = fos.find { |l| l["fosLabel"] == hsh["__content__"] || "FOS: " + l["fosLabel"] == hsh["__content__"] || l["fosLabel"] == hsh["subject"]}
 
       if subject
-        return [{ 
-          "subject" => sanitize(hsh["__content__"] || hsh["subject"]), 
+        return [{
+          "subject" => sanitize(hsh["__content__"] || hsh["subject"]),
           "subjectScheme" => hsh["subjectScheme"],
-          "schemeUri" => hsh["schemeURI"], 
-          "valueUri" => hsh["valueURI"], 
+          "schemeUri" => hsh["schemeURI"],
+          "valueUri" => hsh["valueURI"],
+          "classificationCode" => hsh["classificationCode"],
           "lang" => hsh["lang"] }.compact,
-        {  
+        {
           "subject" => "FOS: " + subject["fosLabel"],
           "subjectScheme" => "Fields of Science and Technology (FOS)",
-          "schemeUri" => "http://www.oecd.org/science/inno/38235147.pdf" }.compact] 
+          "schemeUri" => "http://www.oecd.org/science/inno/38235147.pdf" }.compact]
       end
 
       # if not found, look in Fields of Research (Australian and New Zealand Standard Research Classification)
@@ -1210,8 +1325,8 @@ module Bolognese
       # try to extract forId
       if hsh["subjectScheme"] == "FOR"
         for_id = hsh["__content__"].to_s.split(" ").first || hsh["subject"].to_s.split(" ").first
-        for_id = for_id.rjust(6, "0")        
-        
+        for_id = for_id.rjust(6, "0")
+
         subject = for_fields.find { |l| l["forId"] == for_id } ||
                   for_disciplines.find { |l| l["forId"] == for_id[0..3] }
       else
@@ -1220,11 +1335,12 @@ module Bolognese
       end
 
       if subject
-        [{ 
-          "subject" => sanitize(hsh["__content__"] || hsh["subject"]), 
+        [{
+          "subject" => sanitize(hsh["__content__"] || hsh["subject"]),
           "subjectScheme" => hsh["subjectScheme"],
-          "schemeUri" => hsh["schemeURI"], 
-          "valueUri" => hsh["valueURI"], 
+          "classificationCode" => hsh["classificationCode"],
+          "schemeUri" => hsh["schemeURI"],
+          "valueUri" => hsh["valueURI"],
           "lang" => hsh["lang"] }.compact,
         {
           "subject" => "FOS: " + subject["fosLabel"],
@@ -1232,12 +1348,13 @@ module Bolognese
           "schemeUri" => "http://www.oecd.org/science/inno/38235147.pdf"
         }]
       else
-        [{ 
-          "subject" => sanitize(hsh["__content__"] || hsh["subject"]), 
+        [{
+          "subject" => sanitize(hsh["__content__"] || hsh["subject"]),
           "subjectScheme" => hsh["subjectScheme"],
-          "schemeUri" => hsh["schemeURI"], 
-          "valueUri" => hsh["valueURI"], 
-          "lang" => hsh["lang"] }.compact]  
+          "classificationCode" => hsh["classificationCode"],
+          "schemeUri" => hsh["schemeURI"],
+          "valueUri" => hsh["valueURI"],
+          "lang" => hsh["lang"] }.compact]
       end
     end
   end