dor-services 8.6.0 → 9.0.0

data/lib/dor/datastreams/simple_dublin_core_ds.rb

@@ -1,59 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  class SimpleDublinCoreDs < ActiveFedora::OmDatastream
-    set_terminology do |t|
-      t.root(
-        path: 'dc',
-        xmlns: 'http://www.openarchives.org/OAI/2.0/oai_dc/',
-        schema: 'http://cosimo.stanford.edu/standards/oai_dc/v2/oai_dc.xsd',
-        namespace_prefix: 'oai_dc',
-        index_as: [:not_searchable]
-      )
-      t.title(index_as: %i[stored_sortable stored_searchable], xmlns: 'http://purl.org/dc/elements/1.1/', namespace_prefix: 'dc')
-      t.creator(index_as: %i[stored_sortable stored_searchable], xmlns: 'http://purl.org/dc/elements/1.1/', namespace_prefix: 'dc')
-      t.identifier(index_as: %i[symbol stored_searchable], xmlns: 'http://purl.org/dc/elements/1.1/', namespace_prefix: 'dc')
-    end
-
-    def self.xml_template
-      builder = Nokogiri::XML::Builder.new do |xml|
-        xml['oai_dc'].dc(
-          'xmlns:oai_dc' => 'http://www.openarchives.org/OAI/2.0/oai_dc/',
-          'xmlns:dc' => 'http://purl.org/dc/elements/1.1/',
-          'xmlns:xsi' => 'http://www.w3.org/2001/XMLSchema-instance',
-          'xsi:schemaLocation' => 'http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd'
-        )
-      end
-
-      builder.doc
-    end
-
-    def to_solr(solr_doc = {}, *args)
-      # There are a whole bunch of namespace-related things that can go
-      # wrong with this terminology. Until it's fixed in OM, ignore them all.
-
-      doc = super solr_doc, *args
-
-      add_solr_value(doc, 'dc_title',   title.first,   :string, %i[stored_sortable stored_searchable])
-      add_solr_value(doc, 'dc_creator', creator.first, :string, %i[stored_sortable stored_searchable])
-
-      identifiers = {}
-
-      identifier.each { |i| ns, val = i.split(':'); identifiers[ns] ||= val }
-
-      identifiers.each do |ns, val|
-        add_solr_value(doc, "dc_identifier_#{ns}", val, :string, %i[stored_sortable stored_searchable])
-      end
-
-      doc
-    rescue Exception => e
-      warn "ERROR in SimpleDublinCoreDs to_solr()! #{e}"
-      solr_doc
-    end
-
-    # maintain AF < 8 indexing behavior
-    def prefix
-      ''
-    end
-  end
-end

data/lib/dor/datastreams/workflow_definition_ds.rb

@@ -1,71 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  # @deprecated
-  class WorkflowDefinitionDs < ActiveFedora::OmDatastream
-    include SolrDocHelper
-
-    set_terminology do |t|
-      t.root(path: 'workflow-def', index_as: [:not_searchable])
-      t.process(index_as: [:not_searchable])
-    end
-
-    define_template :process do |builder, workflow, attrs|
-      prereqs = attrs.delete('prerequisite')
-      prereqs = prereqs.split(/\s*,\s*/) if prereqs.is_a?(String)
-      attrs.keys.each { |k| attrs[k.to_s.dasherize.to_sym] = attrs.delete(k) }
-      builder.process(attrs) do |node|
-        Array(prereqs).each do |prereq|
-          (repo, wf, prereq_name) = prereq.split(/:/)
-          if prereq_name.nil?
-            prereq_name = repo
-            repo = nil
-          end
-          if repo == workflow.repo && wf = workflow.name
-            repo = nil
-            wf = nil
-          end
-          attrs = repo.nil? && wf.nil? ? {} : { repository: repo, workflow: wf }
-          node.prereq(attrs) { node.text prereq_name }
-        end
-      end
-    end
-
-    def self.xml_template
-      Nokogiri::XML('<workflow-def/>')
-    end
-
-    def add_process(attributes)
-      ng_xml_will_change!
-      add_child_node(ng_xml.at_xpath('/workflow-def'), :process, self, attributes)
-    end
-
-    def processes
-      ng_xml.xpath('/workflow-def/process').collect do |node|
-        Workflow::Process.new(repo, name, node)
-      end.sort { |a, b| (a.sequence || 0) <=> (b.sequence || 0) }
-    end
-
-    def name
-      ng_xml.at_xpath('/workflow-def/@id').to_s
-    end
-
-    def repo
-      ng_xml.at_xpath('/workflow-def/@repository').to_s
-    end
-
-    def to_solr(solr_doc = {}, *args)
-      solr_doc = super(solr_doc, *args)
-      add_solr_value(solr_doc, 'workflow_name', name, :symbol, [:symbol])
-      processes.each do |p|
-        add_solr_value(solr_doc, 'process', "#{p.name}|#{p.label}", :symbol, [:displayable])
-      end
-      solr_doc
-    end
-
-    # maintain AF < 8 indexing behavior
-    def prefix
-      ''
-    end
-  end
-end

data/lib/dor/datastreams/workflow_ds.rb

@@ -1,20 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  # Represents the datastream that just holds the location of the workflow service
-  class WorkflowDs < ActiveFedora::Datastream
-    before_save :build_location
-
-    # Called before saving, but after a pid has been assigned
-    def build_location
-      return unless new?
-
-      self.dsLocation = File.join(Dor::Config.workflow.url, "dor/objects/#{pid}/workflows")
-    end
-
-    # Called by rubydora. This lets us customize the mime-type
-    def self.default_attributes
-      super.merge(mimeType: 'application/xml')
-    end
-  end
-end

data/lib/dor/indexers/composite_indexer.rb

@@ -1,27 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  class CompositeIndexer
-    attr_reader :indexers
-    def initialize(*indexers)
-      @indexers = indexers
-    end
-
-    def new(resource:)
-      Instance.new(indexers, resource: resource)
-    end
-
-    class Instance
-      attr_reader :indexers, :resource
-      def initialize(indexers, resource:)
-        @resource = resource
-        @indexers = indexers.map { |i| i.new(resource: resource) }
-      end
-
-      # @return [Hash] the merged solr document for all the sub-indexers
-      def to_solr
-        indexers.map(&:to_solr).inject({}, &:merge)
-      end
-    end
-  end
-end

data/lib/dor/indexers/data_indexer.rb

@@ -1,24 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  # Indexing provided by ActiveFedora
-  class DataIndexer
-    include ActiveFedora::Indexing
-
-    attr_reader :resource
-    def initialize(resource:)
-      @resource = resource
-    end
-
-    # we need to override this until https://github.com/samvera/active_fedora/pull/1371
-    # has been released
-    def to_solr(solr_doc = {}, opts = {})
-      super.tap do |doc|
-        doc['active_fedora_model_ssi'] = has_model
-      end
-    end
-
-    delegate :create_date, :modified_date, :state, :pid, :inner_object,
-             :datastreams, :relationships, :has_model, to: :resource
-  end
-end

data/lib/dor/indexers/describable_indexer.rb

@@ -1,60 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  class DescribableIndexer
-    attr_reader :resource
-    def initialize(resource:)
-      @resource = resource
-    end
-
-    # @return [Hash] the partial solr document for describable concerns
-    def to_solr
-      add_metadata_format_to_solr_doc.merge(add_mods_to_solr_doc)
-    end
-
-    def add_metadata_format_to_solr_doc
-      { 'metadata_format_ssim' => 'mods' }
-    end
-
-    def add_mods_to_solr_doc
-      solr_doc = {}
-      mods_sources = {
-        sw_title_display: %w(sw_display_title_tesim),
-        main_author_w_date: %w(sw_author_ssim sw_author_tesim),
-        sw_sort_author: %w(sw_author_sort_ssi),
-        sw_language_facet: %w(sw_language_ssim sw_language_tesim),
-        sw_genre: %w(sw_genre_ssim sw_genre_tesim),
-        format_main: %w(sw_format_ssim sw_format_tesim),
-        topic_facet: %w(sw_topic_ssim sw_topic_tesim),
-        era_facet: %w(sw_subject_temporal_ssim sw_subject_temporal_tesim),
-        geographic_facet: %w(sw_subject_geographic_ssim sw_subject_geographic_tesim),
-        %i[term_values typeOfResource] => %w(mods_typeOfResource_ssim mods_typeOfResource_tesim),
-        pub_year_sort_str: %w(sw_pub_date_sort_ssi),
-        pub_year_int: %w(sw_pub_date_sort_isi),
-        pub_year_display_str: %w(sw_pub_date_facet_ssi)
-      }
-
-      mods_sources.each_pair do |meth, solr_keys|
-        vals = meth.is_a?(Array) ? resource.stanford_mods.send(meth.shift, *meth) : resource.stanford_mods.send(meth)
-
-        next if vals.nil? || (vals.respond_to?(:empty?) && vals.empty?)
-
-        solr_keys.each do |key|
-          solr_doc[key] ||= []
-          solr_doc[key].push(*vals)
-        end
-        # asterisk to avoid multi-dimensional array: push values, not the array
-      end
-
-      # convert multivalued fields to single value
-      %w(sw_pub_date_sort_ssi sw_pub_date_sort_isi sw_pub_date_facet_ssi).each do |key|
-        solr_doc[key] = solr_doc[key].first unless solr_doc[key].nil?
-      end
-      # some fields get explicit "(none)" placeholder values, mostly for faceting
-      %w(sw_language_tesim sw_genre_tesim sw_format_tesim).each do |key|
-        solr_doc[key] = ['(none)'] if solr_doc[key].nil? || solr_doc[key].empty?
-      end
-      solr_doc
-    end
-  end
-end

data/lib/dor/indexers/editable_indexer.rb

@@ -1,25 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  class EditableIndexer
-    include SolrDocHelper
-
-    attr_reader :resource
-    def initialize(resource:)
-      @resource = resource
-    end
-
-    def to_solr
-      {}.tap do |solr_doc|
-        add_solr_value(solr_doc, 'default_rights', default_rights_for_indexing, :string, [:symbol])
-        add_solr_value(solr_doc, 'agreement', resource.agreement, :string, [:symbol]) if resource.agreement_object
-        add_solr_value(solr_doc, 'default_use_license_machine', resource.use_license, :string, [:stored_sortable])
-      end
-    end
-
-    # @return [String] A description of the rights defined in the default object rights datastream. Can be 'Stanford', 'World', 'Dark' or 'None'
-    def default_rights_for_indexing
-      RightsMetadataDS::RIGHTS_TYPE_CODES.fetch(resource.default_rights, 'Unrecognized default rights value')
-    end
-  end
-end

data/lib/dor/indexers/identifiable_indexer.rb

@@ -1,102 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  class IdentifiableIndexer
-    include SolrDocHelper
-
-    INDEX_VERSION_FIELD = 'dor_services_version_ssi'
-    NS_HASH = { 'hydra' => 'http://projecthydra.org/ns/relations#',
-                'fedora' => 'info:fedora/fedora-system:def/relations-external#',
-                'rdf' => 'http://www.w3.org/1999/02/22-rdf-syntax-ns#' }.freeze
-
-    attr_reader :resource
-    def initialize(resource:)
-      @resource = resource
-    end
-
-    ## Module-level variables, shared between ALL mixin includers (and ALL *their* includers/extenders)!
-    ## used for caching found values
-    @@collection_hash = {}
-    @@apo_hash = {}
-
-    # @return [Hash] the partial solr document for identifiable concerns
-    def to_solr
-      solr_doc = {}
-      solr_doc[INDEX_VERSION_FIELD] = Dor::VERSION
-      solr_doc['indexed_at_dtsi'] = Time.now.utc.xmlschema
-      resource.datastreams.values.each do |ds|
-        # This is used to draw the table of datastreams in Argo
-        add_solr_value(solr_doc, 'ds_specs', ds.datastream_spec_string, :string, [:symbol]) unless ds.new?
-      end
-
-      add_solr_value(solr_doc, 'title_sort', resource.label, :string, [:stored_sortable])
-
-      rels_doc = Nokogiri::XML(resource.datastreams['RELS-EXT'].content)
-      apos = rels_doc.search('//rdf:RDF/rdf:Description/hydra:isGovernedBy', NS_HASH)
-      collections = rels_doc.search('//rdf:RDF/rdf:Description/fedora:isMemberOfCollection', NS_HASH)
-      solrize_related_obj_titles(solr_doc, apos, @@apo_hash, 'apo_title', 'nonhydrus_apo_title', 'hydrus_apo_title')
-      solrize_related_obj_titles(solr_doc, collections, @@collection_hash, 'collection_title', 'nonhydrus_collection_title', 'hydrus_collection_title')
-      solr_doc['public_dc_relation_tesim'] ||= solr_doc['collection_title_tesim'] if solr_doc['collection_title_tesim']
-      solr_doc['metadata_source_ssi'] = identity_metadata_source
-      solr_doc
-    end
-
-    # @return [String] calculated value for Solr index
-    def identity_metadata_source
-      if resource.identityMetadata.otherId('catkey').first ||
-         resource.identityMetadata.otherId('barcode').first
-        'Symphony'
-      else
-        'DOR'
-      end
-    end
-
-    # Clears out the cache of items. Used primarily in testing.
-    def self.reset_cache!
-      @@collection_hash = {}
-      @@apo_hash = {}
-    end
-
-    private
-
-    def solrize_related_obj_titles(solr_doc, relationships, title_hash, union_field_name, nonhydrus_field_name, hydrus_field_name)
-      # TODO: if you wanted to get a little fancier, you could also solrize a 2 level hierarchy and display using hierarchial facets, like
-      # ["SOURCE", "SOURCE : TITLE"] (e.g. ["Hydrus", "Hydrus : Special Collections"], see (exploded) tags in IdentityMetadataDS#to_solr).
-      title_type = :symbol # we'll get an _ssim because of the type
-      title_attrs = [:stored_searchable] # we'll also get a _tesim from this attr
-      relationships.each do |rel_node|
-        rel_druid = rel_node['rdf:resource']
-        next unless rel_druid # TODO: warning here would also be useful
-
-        rel_druid = rel_druid.gsub('info:fedora/', '')
-
-        # populate cache if necessary
-        unless title_hash.key?(rel_druid)
-          begin
-            related_obj = Dor.find(rel_druid)
-            related_obj_title = related_obj_display_title(related_obj, rel_druid)
-            is_from_hydrus = (related_obj&.tags&.include?('Project : Hydrus'))
-            title_hash[rel_druid] = { 'related_obj_title' => related_obj_title, 'is_from_hydrus' => is_from_hydrus }
-          rescue ActiveFedora::ObjectNotFoundError
-            # This may happen if the given APO or Collection does not exist (bad data)
-            title_hash[rel_druid] = { 'related_obj_title' => rel_druid, 'is_from_hydrus' => false }
-          end
-        end
-
-        # cache should definitely be populated, so just use that to write solr field
-        if title_hash[rel_druid]['is_from_hydrus']
-          add_solr_value(solr_doc, hydrus_field_name, title_hash[rel_druid]['related_obj_title'], title_type, title_attrs)
-        else
-          add_solr_value(solr_doc, nonhydrus_field_name, title_hash[rel_druid]['related_obj_title'], title_type, title_attrs)
-        end
-        add_solr_value(solr_doc, union_field_name, title_hash[rel_druid]['related_obj_title'], title_type, title_attrs)
-      end
-    end
-
-    def related_obj_display_title(related_obj, default_title)
-      return default_title unless related_obj
-
-      related_obj.full_title || default_title
-    end
-  end
-end

data/lib/dor/indexers/process_indexer.rb

@@ -1,58 +0,0 @@
-# frozen_string_literal: true
-
-module Dor
-  # Indexes the process for a workflow
-  class ProcessIndexer
-    ERROR_OMISSION = '... (continued)'
-    private_constant :ERROR_OMISSION
-
-    # see https://lucene.apache.org/core/7_3_1/core/org/apache/lucene/util/BytesRefHash.MaxBytesLengthExceededException.html
-    MAX_ERROR_LENGTH = 32_768 - 2 - ERROR_OMISSION.length
-    private_constant :MAX_ERROR_LENGTH
-
-    # @param [WorkflowSolrDocument] solr_doc
-    # @param [String] workflow_name
-    # @param [Dor::Workflow::Response::Process] process
-    def initialize(solr_doc:, workflow_name:, process:)
-      @solr_doc = solr_doc
-      @workflow_name = workflow_name
-      @process = process
-    end
-
-    # @return [Hash] the partial solr document for the workflow document
-    def to_solr
-      return unless status
-
-      # add a record of the robot having operated on this item, so we can track robot activity
-      solr_doc.add_process_time(workflow_name, name, Time.parse(process.datetime)) if has_time?
-
-      index_error_message
-
-      # workflow name, process status then process name
-      solr_doc.add_wsp("#{workflow_name}:#{status}", "#{workflow_name}:#{status}:#{name}")
-
-      # workflow name, process name then process status
-      solr_doc.add_wps("#{workflow_name}:#{name}", "#{workflow_name}:#{name}:#{status}")
-
-      # process status, workflowname then process name
-      solr_doc.add_swp(process.status.to_s, "#{status}:#{workflow_name}", "#{status}:#{workflow_name}:#{name}")
-    end
-
-    private
-
-    attr_reader :process, :workflow_name, :solr_doc
-    delegate :status, :name, :state, :error_message, :datetime, to: :process
-
-    def has_time?
-      datetime && (status == 'completed' || status == 'error')
-    end
-
-    # index the error message without the druid so we hopefully get some overlap
-    # truncate to avoid org.apache.lucene.util.BytesRefHash$MaxBytesLengthExceededException
-    def index_error_message
-      return unless error_message
-
-      solr_doc.error = "#{workflow_name}:#{name}:#{error_message}".truncate(MAX_ERROR_LENGTH, omission: ERROR_OMISSION)
-    end
-  end
-end