uc3-dmp-id 0.1.78 → 0.1.79

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 10b0e02321270e309ee7352e0adfc06f9f86d5b57397dc2d34cdd373aaacaadf
-  data.tar.gz: 2985739c95e04e3d515cfce61f6ab61ec2e99bbb8edce925c8222b6b194216a6
+  metadata.gz: d05302ee8b8190c77e7efc0ad602ad0de2a74739b1653672d707b1fda95eec16
+  data.tar.gz: 3cca6911d51d58ff974308f8647bab075fb9bf717092bce45f3cbe2a151c7712
 SHA512:
-  metadata.gz: f0594eb5e5faa1c214f953747a649dd5ff932eb7c554eada2db2e6e5dcc81c52565942768132c6d781000a84c55ccfa606829dd68f9775cd0f104b5070927c5b
-  data.tar.gz: 50142c1ea8e666446ca51075ed1ddfa73a46e8ef68df87e64ca75e5da5bdfc1470fded80c52fe0c6524e65b1c365dce57146535bf744552da77273d9b3881074
+  metadata.gz: e19a514c8233bdf9abc32c375b636611181738f6f64818431d0b3e26b1fd40c3817b5dcf1cfc8e1a270e3c521ad67636dd50baaec7638af9a162ebb864b26975
+  data.tar.gz: 3b09fdc606348c0c3dddcc7536563920c2971d475981c1b11a89a7147157db712d416d52eafec3689ec0e004be3d845acdf83cbb2113b2df08aca58936421915

data/lib/uc3-dmp-id/finder.rb

@@ -17,22 +17,41 @@ module Uc3DmpId
     MSG_MISSING_PROV_ID = 'No Provenance identifier was provided. \
                            Expected: `{ "dmp_id": { "identifier": "value", "type": "value" }`'
 
+    ORCID_DOMAIN = 'https://orcid.org/'
+    ROR_DOMAIN = 'https://ror.org/'
+    DOI_DOMAIN = 'https://doi.org/'
+    SORT_OPTIONS = %w[title modified]
+    SORT_DIRECTIONS = %w[asc desc]
+    MAX_PAGE_SIZE = 100
+    DEFAULT_PAGE_SIZE = 25
+    DEFAULT_SORT_OPTION = 'modified'
+    DEFAULT_SORT_DIR = 'desc'
+
     class << self
       # TODO: Replace this with ElasticSearch
       def search_dmps(args:, logger: nil)
-        client = Uc3DmpDynamo::Client.new
-        return _by_owner(owner_org: args['owner_orcid'], client:, logger:) unless args['owner_orcid'].nil?
+        # Fetch the DMPs for each of the possible filter options
+        client = Uc3DmpDynamo::Client.new(table: ENV['DYNAMO_INDEX_TABLE'])
+        owner = args['owner']
+        org = args['org']
+        funder = args['funder']
+
+        owner_pks = owner.nil? ? [] : _by_owner(owner: owner, client:, logger:)
+        org_pks = org.nil? ? [] : _by_org(org: org, client:, logger:)
+        funder_pks = funder.nil? ? [] : _by_funder(funder: funder, client:, logger:)
+        # pks = [owner_pks, org_pks, funder_pks].reject(&:empty?)
+        logger&.debug(
+          message: 'PKs found',
+          details: { owner: owner_pks, org: org_pks, funder: funder_pks }
+        )
+        # return [] if pks.nil? || pks.empty?
 
-        unless args['owner_org_ror'].nil?
-          return _by_owner_org(owner_org: args['owner_org_ror'], client:,
-                               logger:)
-        end
-        unless args['modification_day'].nil?
-          return _by_mod_day(day: args['modification_day'], client:,
-                             logger:)
-        end
+        # Only use the DMPs that fit all of the filter criteria
+        # dmps = pks.reduce(:&).flatten.uniq
+        # return [] if dmps.nil? || dmps.empty?
+
+        [owner_pks, org_pks, funder_pks].flatten.uniq
 
-        []
       end
 
       # Find a DMP based on the contents of the incoming JSON
@@ -134,74 +153,66 @@ module Uc3DmpId
 
       private
 
-      # Fetch the DMP IDs for the specified owner's ORCID (the owner is the :dmphub_owner_id on the DMP ID record)
-      def _by_owner(owner_id:, client: nil, logger: nil)
-        regex = /^([0-9A-Z]{4}-){3}[0-9A-Z]{4}$/
-        raise FinderError, MSG_INVALID_OWNER_ID if owner_id.nil? || (owner_id.to_s =~ regex).nil?
+      # Fetch the DMP IDs for the specified person's ORCID (or email)
+      def _by_owner(owner:, client: nil, logger: nil)
+        orcid_regex = /^([0-9a-zA-Z]{4}-){3}[0-9a-zA-Z]{4}$/
+        email_regex = /^[^\s@]+@[^\s@]+\.[^\s@]+$/
+        orcid = owner.to_s.strip
+        return [] if (orcid =~ orcid_regex).nil? && (orcid =~ email_regex).nil?
 
-        args = {
-          index_name: 'dmphub_owner_id_gsi',
-          key_conditions: {
-            dmphub_owner_id: {
-              attribute_value_list: [
-                "http://orcid.org/#{owner_id}",
-                "https://orcid.org/#{owner_id}"
-              ],
-              comparison_operator: 'IN'
-            }
-          },
-          filter_expression: 'SK = :version',
-          expression_attribute_values: { ':version': Helper::DMP_LATEST_VERSION }
-        }
-        logger.info(message: "Querying _by_owner with #{args}") if logger.respond_to?(:info)
-        client = Uc3DmpDynamo::Client.new if client.nil?
-        _process_search_response(response: client.query(args:, logger:))
+        orcid = "#{ORCID_DOMAIN}#{orcid}" unless (orcid =~ orcid_regex).nil?
+        resp = client.get_item(key: { PK: 'PERSON_INDEX', SK: orcid }, logger:)
+        return [] unless resp.is_a?(Hash)
+
+        logger&.debug(message: "DMPs for PERSON #{orcid}", details: resp)
+        resp.fetch('dmps', [])
       end
 
-      # Fetch the DMP IDs for the specified organization/institution (the org is the :dmphub_owner_org
-      # on the DMP ID record)
-      def _by_owner_org(owner_org:, client: nil, logger: nil)
+      # Fetch the DMP IDs for the specified organization/institution
+      def _by_org(org:, client: nil, logger: nil)
         regex = /^[a-zA-Z0-9]+$/
-        raise FinderError, MSG_INVALID_OWNER_ID if owner_org.nil? || (owner_org.to_s.downcase =~ regex).nil?
+        id = "#{ROR_DOMAIN}#{org.strip}" unless (org.to_s =~ regex).nil?
+
+        resp = client.get_item(key: { PK: 'AFFILIATION_INDEX', SK: id }, logger:)
+        return [] unless resp.is_a?(Hash)
+
+        logger&.debug(message: "DMPs for AFFILIATION #{id}", details: resp)
+        resp.fetch('dmps', [])
+      end
 
+      # Fetch the DMP IDs for the specified funder
+      def _by_funder(funder:, client: nil, logger: nil)
+        regex = /^[a-zA-Z0-9]+$/
+        id = "#{ROR_DOMAIN}/#{funder.strip}" unless (funder.to_s =~ regex).nil?
+        id = "#{DOI_DOMAIN}#{org.strip}" if id.nil? && !(org.to_s =~ Helper::DOI_REGEX).nil?
+
+        resp = client.get_item(key: { PK: 'FUNDER_INDEX', SK: id }, logger:)
+        return [] unless resp.is_a?(Hash)
+
+        logger&.debug(message: "DMPs for FUNDER #{id}", details: resp)
+        resp.fetch('dmps', [])
+      end
+
+      # Fetch the DMP IDs that are marked as featured
+      def _by_featured(client: nil, logger: nil)
         args = {
-          index_name: 'dmphub_owner_org_gsi',
-          key_conditions: {
-            dmphub_owner_org: {
-              attribute_value_list: [
-                "https://ror.org/#{owner_org.to_s.downcase}",
-                "http://ror.org/#{owner_org.to_s.downcase}"
-              ],
-              comparison_operator: 'IN'
-            }
-          },
-          filter_expression: 'SK = :version',
-          expression_attribute_values: { ':version': Helper::DMP_LATEST_VERSION }
+          filter_expression: 'featured = :featured AND SK = :sk',
+          expression_attribute_values: { ':sk': 'METADATA', ':featured': 1 }
         }
-        logger.info(message: "Querying _by_owner_org with #{args}") if logger.respond_to?(:info)
+        logger&.debug(message: 'Fetch relevant DMPs _by_featured - scan args', details: args)
         client = Uc3DmpDynamo::Client.new if client.nil?
-        _process_search_response(response: client.query(args:, logger:))
+        _process_search_response(response: client.scan(args:))
       end
 
-      # Fetch the DMP IDs modified on the specified date (the date is the :dmphub_modification_day on the DMP ID record)
-      def _by_mod_day(day:, client: nil, logger: nil)
-        regex = /^[0-9]{4}(-[0-9]{2}){2}/
-        raise FinderError, MSG_INVALID_OWNER_ID if day.nil? || (day.to_s =~ regex).nil?
-
+      # Return all of the publicly visible DMPs
+      def _publicly_visible(client: nil, logger: nil)
         args = {
-          index_name: 'dmphub_modification_day_gsi',
-          key_conditions: {
-            dmphub_modification_day: {
-              attribute_value_list: [day.to_s],
-              comparison_operator: 'IN'
-            }
-          },
-          filter_expression: 'SK = :version',
-          expression_attribute_values: { ':version': Helper::DMP_LATEST_VERSION }
+          filter_expression: 'visibility = :visibility AND SK = :sk',
+          expression_attribute_values: { ':sk': 'METADATA', ':visibility': 'public' }
         }
-        logger.info(message: "Querying _by_mod_day with #{args}") if logger.respond_to?(:info)
+        logger&.debug(message: 'Fetch relevant DMPs _publicly_visible - scan args', details: args)
         client = Uc3DmpDynamo::Client.new if client.nil?
-        _process_search_response(response: client.query(args:, logger:))
+        _process_search_response(response: client.scan(args:))
       end
 
       # Transform the search results so that we do not include any of the DMPHub specific metadata
@@ -212,8 +223,8 @@ module Uc3DmpId
           next if item.nil?
 
           dmp = item['dmp'].nil? ? JSON.parse({ dmp: item }.to_json) : item
-          dmp = _remove_narrative_if_private(json: dmp)
-          Helper.cleanse_dmp_json(json: dmp)
+          # dmp = _remove_narrative_if_private(json: dmp)
+          # Helper.cleanse_dmp_json(json: dmp)
         end
         results.compact.uniq
       end

data/lib/uc3-dmp-id/updater.rb

@@ -16,9 +16,9 @@ module Uc3DmpId
       def update(provenance:, p_key:, json: {}, logger: nil)
         raise UpdaterError, Helper::MSG_DMP_INVALID_DMP_ID unless p_key.is_a?(String) && !p_key.strip.empty?
 
-        payload = Helper.parse_json(json:).fetch('dmp', {})
+        mods = Helper.parse_json(json:).fetch('dmp', {})
         p_key = Helper.append_pk_prefix(p_key:)
-        logger.debug(message: "Incoming modifications for PK #{p_key}", details: payload) if logger.respond_to?(:debug)
+        logger.debug(message: "Incoming modifications for PK #{p_key}", details: mods) if logger.respond_to?(:debug)
 
         # Fetch the latest version of the DMP ID
         client = Uc3DmpDynamo::Client.new
@@ -28,30 +28,29 @@ module Uc3DmpId
 
         # Verify that the DMP ID is updateable with the info passed in
         errs = _updateable?(provenance:, p_key:, latest_version: latest_version['dmp'],
-                            mods: payload['dmp'])
+                            mods: mods['dmp'])
         logger.error(message: errs.join(', ')) if logger.respond_to?(:error) && errs.is_a?(Array) && errs.any?
         raise UpdaterError, errs if errs.is_a?(Array) && errs.any?
         # Don't continue if nothing has changed!
-        raise UpdaterError, Helper::MSG_NO_CHANGE if Helper.eql?(dmp_a: latest_version, dmp_b: payload)
+        raise UpdaterError, Helper::MSG_NO_CHANGE if Helper.eql?(dmp_a: latest_version, dmp_b: mods)
 
         # Version the DMP ID record (if applicable).
         owner = latest_version['dmphub_provenance_id']
         updater = provenance['PK']
         version = Versioner.generate_version(client:, latest_version:, owner:,
                                              updater:, logger:)
-        logger&.debug(message: 'New Version', details: version)
         raise UpdaterError, Helper::MSG_DMP_UNABLE_TO_VERSION if version.nil?
         # Bail if the system trying to make the update is not the creator of the DMP ID
         raise UpdaterError, Helper::MSG_DMP_FORBIDDEN if owner != updater
 
         # Handle any changes to the dmphub_modifications section
-        version = _process_harvester_mods(client:, p_key:, json: payload, version:, logger:)
-        logger&.debug(message: 'Version after process_harvester_mods', details: version)
+        version = _process_harvester_mods(client:, p_key:, json: version, logger:)
 
-        # Remove the version info any any lingering modification blocks
+        # Remove the version info because we don't want to save it on the record
         version.delete('dmphub_versions')
-        version.delete('dmphub_modifications')
 
+        # Splice the assertions
+        version = _process_modifications(owner:, updater:, version:, mods:, logger:)
         # Set the :modified timestamps
         now = Time.now.utc
         version['modified'] = now.iso8601
@@ -68,9 +67,9 @@ module Uc3DmpId
         logger.info(message: "Updated DMP ID: #{p_key}") if logger.respond_to?(:debug)
 
         # Append the :dmphub_versions Array
-        out = JSON.parse({ dmp: version }.to_json)
-        out = Versioner.append_versions(p_key:, dmp: out, client:, logger:)
-        Helper.cleanse_dmp_json(json: out)
+        json = JSON.parse({ dmp: version }.to_json)
+        json = Versioner.append_versions(p_key:, dmp: json, client:, logger:)
+        Helper.cleanse_dmp_json(json:)
       end
       # rubocop:enable Metrics/AbcSize, Metrics/MethodLength
       # rubocop:enable Metrics/CyclomaticComplexity, Metrics/PerceivedComplexity
@@ -131,6 +130,24 @@ module Uc3DmpId
       end
       # rubocop:enable Metrics/AbcSize
 
+      # rubocop:disable Metrics/AbcSize, Metrics/CyclomaticComplexity, Metrics/PerceivedComplexity
+      def _process_modifications(owner:, updater:, version:, mods:, logger: nil)
+        return version unless mods.is_a?(Hash) && !updater.nil?
+        return mods unless version.is_a?(Hash) && !owner.nil?
+
+        logger.debug(message: 'Modifications before merge.', details: mods) if logger.respond_to?(:debug)
+        keys_to_retain = version.keys.select do |key|
+          (key.start_with?('dmphub_') && !%w[dmphub_modifications dmphub_versions].include?(key)) ||
+            key.start_with?('PK') || key.start_with?('SK') || key.start_with?('dmproadmap_related_identifiers')
+        end
+        keys_to_retain.each do |key|
+          mods[key] = version[key]
+        end
+        logger.debug(message: 'Modifications after merge.', details: mods) if logger.respond_to?(:debug)
+        mods
+      end
+      # rubocop:enable Metrics/AbcSize, Metrics/CyclomaticComplexity, Metrics/PerceivedComplexity
+
       # Once the DMP has been updated, we need to update it's DOI metadata
       # -------------------------------------------------------------------------
       # rubocop:disable Metrics/AbcSize, Metrics/MethodLength
@@ -174,8 +191,7 @@ module Uc3DmpId
       # rubocop:enable Metrics/PerceivedComplexity, Metrics/CyclomaticComplexity
 
       # Fetch any Harvester modifications to the JSON
-      def _process_harvester_mods(client:, p_key:, json:, version:, logger: nil)
-        logger&.debug(message: 'Incoming modifications', details: json)
+      def _process_harvester_mods(client:, p_key:, json:, logger: nil)
         return json if json.fetch('dmphub_modifications', []).empty?
 
         # Fetch the `"SK": "HARVESTER_MODS"` record
@@ -185,54 +201,22 @@ module Uc3DmpId
         )
         return json unless resp.is_a?(Hash) && resp['related_works'].is_a?(Hash)
 
-        logger&.debug(message: 'Original HARVESTER_MODS record', details: resp)
         # The `dmphub_modifications` array will ONLY ever have things the harvester mods know about
         # so just find them and update the status accordingly
-        original = resp.dup
-        json['dmproadmap_related_identifiers'] = [] if json['dmproadmap_related_identifiers'].nil?
-
+        mods = resp.dup
         json['dmphub_modifications'].each do |entry|
           next if entry.fetch('dmproadmap_related_identifiers', []).empty?
 
           entry['dmproadmap_related_identifiers'].each do |related|
-            # Detrmine if the HARVESTER_MODS record even knows about the mod
-            related_id = related.respond_to?(:identifier) ? related.identifier : related['identifier']
-            related_domain = related.respond_to?(:domain) ? related.domain : related['domain']
-            key = "#{related_domain.end_with?('/') ? related_domain : "#{related_domain}/"}#{related_id}"
-            key_found = original['related_works'].has_key?(key)
-            logger&.debug(message: "No matching HARVEST_MOD found for #{key}") unless key_found
-            next unless key_found
-
-            # Update the status in the HARVESTER_MODS record
-            logger&.debug(message: "Updating status for #{key} from #{original['related_works'][key]['status']} to #{related['status']}")
-            original['related_works'][key]['status'] = related['status']
-
-            existing = version['dmproadmap_related_identifiers'].select do |ri|
-              ri['identifier'] == key
-            end
-
-            # Add it if it was approved and doesn't exist in dmproadmap_related_identifiers
-            if related['status'] == 'approved' && existing.empty?
-              version['dmproadmap_related_identifiers'] << {
-                identifier: key,
-                work_type: related['work_type'],
-                type: related['type'],
-                descriptor: related['descriptor'],
-                citation: related['citation']
-              }
-            elsif related['status'] == 'rejected' && existing.any?
-              # otherwise remove it
-              version['dmproadmap_related_identifiers'] = version['dmproadmap_related_identifiers'].reject { |ri| ri == existing.first }
-            end
+            next if mods['related_works'][related.identifier].nil?
+
+            mods['related_works'][related.identifier]['status'] = related['status']
           end
         end
 
-        logger&.debug(message: 'Updating HARVESTER_MODS with:', details: original)
-        resp = client.put_item(json: original, logger:)
-        logger&.error(message: 'Unable to update HARVESTER_MODS', details: original) if resp.nil?
-
-        logger&.debug(message: 'Returning updated VERSION:', details: version)
-        version
+        client.put_item(json: mods, logger:)
+        json.delete('dmphub_modifications')
+        json
       end
     end
   end

data/lib/uc3-dmp-id/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Uc3DmpId
-  VERSION = '0.1.78'
+  VERSION = '0.1.79'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: uc3-dmp-id
 version: !ruby/object:Gem::Version
-  version: 0.1.78
+  version: 0.1.79
 platform: ruby
 authors:
 - Brian Riley
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-07-12 00:00:00.000000000 Z
+date: 2024-07-16 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: json