dor-services 5.2.0 → 5.3.0

data/lib/dor/models/preservable.rb

@@ -5,8 +5,8 @@ module Dor
     extend ActiveSupport::Concern
 
     included do
-      has_metadata :name => "provenanceMetadata", :type => ActiveFedora::OmDatastream, :label => 'Provenance Metadata'
-      has_metadata :name => "technicalMetadata", :type => ActiveFedora::OmDatastream, :label => 'Technical Metadata', :control_group => 'M'
+      has_metadata :name => 'provenanceMetadata', :type => ActiveFedora::OmDatastream, :label => 'Provenance Metadata'
+      has_metadata :name => 'technicalMetadata', :type => ActiveFedora::OmDatastream, :label => 'Technical Metadata', :control_group => 'M'
     end
 
     def build_provenanceMetadata_datastream(workflow_id, event_text)
@@ -15,24 +15,24 @@ module Dor
       ds = datastreams[dsname]
       ds.label = 'Provenance Metadata' unless datastreams.keys.include?(dsname)
       ds.ng_xml = workflow_provenance
-      ds.content=ds.ng_xml.to_s
+      ds.content = ds.ng_xml.to_s
       ds.save
     end
 
-    def build_technicalMetadata_datastream(ds=nil)
+    def build_technicalMetadata_datastream(ds = nil)
       TechnicalMetadataService.add_update_technical_metadata(self)
     end
 
     def sdr_ingest_transfer(agreement_id)
-      SdrIngestService.transfer(self,agreement_id)
+      SdrIngestService.transfer(self, agreement_id)
     end
 
     # @return [Nokogiri::Document]
     def create_workflow_provenance(workflow_id, event_text)
       builder = Nokogiri::XML::Builder.new do |xml|
-        xml.provenanceMetadata(:objectId => self.pid) {
+        xml.provenanceMetadata(:objectId => pid) {
           xml.agent(:name => 'DOR') {
-            xml.what(:object => self.pid) {
+            xml.what(:object => pid) {
               xml.event(:who => "DOR-#{workflow_id}", :when => Time.new.iso8601) {
                 xml.text(event_text)
               }
@@ -42,6 +42,5 @@ module Dor
       end
       builder.doc
     end
-
   end
 end

data/lib/dor/models/processable.rb

@@ -11,9 +11,9 @@ module Dor
       after_initialize :set_workflows_datastream_location
     end
 
-    #verbiage we want to use to describe an item when it has completed a particular step
+    # verbiage we want to use to describe an item when it has completed a particular step
     STATUS_CODE_DISP_TXT = {
-      0 => 'Unknown Status', #if there are no milestones for the current version, someone likely messed up the versioning process.
+      0 => 'Unknown Status', # if there are no milestones for the current version, someone likely messed up the versioning process.
       1 => 'Registered',
       2 => 'In accessioning',
       3 => 'In accessioning (described)',
@@ -25,54 +25,55 @@ module Dor
       9 => 'Opened'
     }
 
-    #milestones from accessioning and the order they happen in
+    # milestones from accessioning and the order they happen in
     STEPS = {
-      'registered' => 1,
-      'submitted' => 2,
-      'described' => 3,
-      'published' => 4,
-      'deposited' => 5,
+      'registered'  => 1,
+      'submitted'   => 2,
+      'described'   => 3,
+      'published'   => 4,
+      'deposited'   => 5,
       'accessioned' => 6,
-      'indexed' => 7,
-      'shelved' => 8,
-      'opened' => 1
+      'indexed'     => 7,
+      'shelved'     => 8,
+      'opened'      => 9
     }
 
     # This is a work-around for some strange logic in ActiveFedora that
     # don't allow self.workflows.new? to work if we load the object using
     # .load_instance_from_solr.
     def set_workflows_datastream_location
-      return if self.respond_to?(:inner_object) && self.inner_object.is_a?(ActiveFedora::SolrDigitalObject)
-      return unless self.workflows.new?
+      return if self.respond_to?(:inner_object) && inner_object.is_a?(ActiveFedora::SolrDigitalObject)
+      return unless workflows.new?
       workflows.mimeType   = 'application/xml'
-      workflows.dsLocation = File.join(Dor::Config.workflow.url,"dor/objects/#{self.pid}/workflows")
+      workflows.dsLocation = File.join(Dor::Config.workflow.url, "dor/objects/#{pid}/workflows")
     end
 
     def empty_datastream?(datastream)
-      if datastream.new?
-        true
-      elsif datastream.class.respond_to?(:xml_template)
+      return true if datastream.new?
+      if datastream.class.respond_to?(:xml_template)
         datastream.content.to_s.empty? || EquivalentXml.equivalent?(datastream.content, datastream.class.xml_template)
       else
         datastream.content.to_s.empty?
       end
     end
 
-    # Takes the name of a datastream, as a string.
     # Tries to find a file for the datastream.
-    # Returns the path to it or nil.
+    # @param [String] datastream name of a datastream
+    # @return [String, nil] path to datastream or nil
     def find_metadata_file(datastream)
       druid = DruidTools::Druid.new(pid, Dor::Config.stacks.local_workspace_root)
-      return druid.find_metadata("#{datastream}.xml")
+      druid.find_metadata("#{datastream}.xml")
     end
 
-    # Takes the name of a datastream, as a string (fooMetadata).
     # Builds that datastream using the content of a file if such a file
     # exists and is newer than the object's current datastream; otherwise,
     # builds the datastream by calling build_fooMetadata_datastream.
+    # @param [String] datastream name of a datastream (e.g. "fooMetadata")
+    # @param [Boolean] force overwrite existing datastream
+    # @param [Boolean] is_required
+    # @return [SomeDatastream]
     def build_datastream(datastream, force = false, is_required = false)
-      # See if the datastream exists as a file and if the file's
-      # timestamp is newer than the datastream's timestamp.
+      # See if datastream exists as a file and if the file's timestamp is newer than datastream's timestamp.
       ds       = datastreams[datastream]
       filename = find_metadata_file(datastream)
       use_file = filename && (ds.createDate.nil? || File.mtime(filename) >= ds.createDate)
@@ -85,15 +86,13 @@ module Dor
       elsif force || empty_datastream?(ds)
         meth = "build_#{datastream}_datastream".to_sym
         if respond_to?(meth)
-          content = self.send(meth, ds)
+          send(meth, ds)
           ds.save unless ds.digital_object.new?
         end
       end
       # Check for success.
-      if is_required && empty_datastream?(ds)
-        raise "Required datastream #{datastream} could not be populated!"
-      end
-      return ds
+      raise "Required datastream #{datastream} could not be populated!" if is_required && empty_datastream?(ds)
+      ds
     end
 
     def cleanup
@@ -101,20 +100,20 @@ module Dor
     end
 
     def milestones
-      Dor::WorkflowService.get_milestones('dor',self.pid)
+      Dor::WorkflowService.get_milestones('dor', pid)
     end
 
     # @return [Hash] including :current_version, :status_code and :status_time
     def status_info
       current_version = '1'
       begin
-        current_version = self.versionMetadata.current_version_id
+        current_version = versionMetadata.current_version_id
       rescue
       end
 
       current_milestones = []
-      #only get steps that are part of accessioning and part of the current version. That can mean they were archived with the current version
-      #number, or they might be active (no version number).
+      # only get steps that are part of accessioning and part of the current version. That can mean they were archived with the current version
+      # number, or they might be active (no version number).
       milestones.each do |m|
         if STEPS.keys.include?(m[:milestone]) && (m[:version].nil? || m[:version] == current_version)
           current_milestones << m unless m[:milestone] == 'registered' && current_version.to_i > 1
@@ -122,57 +121,57 @@ module Dor
       end
 
       status_code = 0
-      status_time = ''
-      #for each milestone in the current version, see if it comes after the current 'last' step, if so, make it the last and record the date/time
+      status_time = nil
+      # for each milestone in the current version, see if it comes after the current 'last' step, if so, make it the last and record the date/time
       current_milestones.each do |m|
-        name = m[:milestone]
-        time = m[:at].utc.xmlschema
-        next unless STEPS.keys.include?(name) && STEPS[name] > status_code
-        status_code = STEPS[name]
-        status_time = time
+        m_name = m[:milestone]
+        m_time = m[:at].utc.xmlschema
+        next unless STEPS.keys.include?(m_name) && (!status_time || m_time > status_time)
+        status_code = STEPS[m_name]
+        status_time = m_time
       end
 
-      return {:current_version => current_version, :status_code => status_code, :status_time => status_time}
+      {:current_version => current_version, :status_code => status_code, :status_time => status_time}
     end
 
     # @param [Boolean] include_time
     # @return [String] single composed status from status_info
-    def status(include_time=false)
-      status_info_hash = status_info()
+    def status(include_time = false)
+      status_info_hash = status_info
       current_version, status_code, status_time = status_info_hash[:current_version], status_info_hash[:status_code], status_info_hash[:status_time]
 
-      #use the translation table to get the appropriate verbage for the latest step
+      # use the translation table to get the appropriate verbage for the latest step
       result = "v#{current_version} #{STATUS_CODE_DISP_TXT[status_code]}"
       result += " #{format_date(status_time)}" if include_time
-      return result
+      result
     end
 
     # return the text translation of the status code, minus any trailing parenthetical explanation
     # e.g. 'In accessioning (described)' and 'In accessioning (described, published)' both come back
     # as 'In accessioning'
     def simplified_status_code_disp_txt(status_code)
-      return STATUS_CODE_DISP_TXT[status_code].gsub(/\(.*\)$/, '').strip
+      STATUS_CODE_DISP_TXT[status_code].gsub(/\(.*\)$/, '').strip
     end
 
-    def to_solr(solr_doc=Hash.new, *args)
+    def to_solr(solr_doc = {}, *args)
       super(solr_doc, *args)
       sortable_milestones = {}
-      current_version='1'
+      current_version = '1'
       begin
-        current_version = self.versionMetadata.current_version_id
+        current_version = versionMetadata.current_version_id
       rescue
       end
-      current_version_num=current_version.to_i
+      current_version_num = current_version.to_i
 
       if self.respond_to?('versionMetadata')
-        #add an entry with version id, tag and description for each version
+        # add an entry with version id, tag and description for each version
         while current_version_num > 0
-          add_solr_value(solr_doc, 'versions', current_version_num.to_s + ';' + self.versionMetadata.tag_for_version(current_version_num.to_s) + ';' + self.versionMetadata.description_for_version(current_version_num.to_s), :string, [:displayable])
+          add_solr_value(solr_doc, 'versions', current_version_num.to_s + ';' + versionMetadata.tag_for_version(current_version_num.to_s) + ';' + versionMetadata.description_for_version(current_version_num.to_s), :string, [:displayable])
           current_version_num -= 1
         end
       end
 
-      self.milestones.each do |milestone|
+      milestones.each do |milestone|
         timestamp = milestone[:at].utc.xmlschema
         sortable_milestones[milestone[:milestone]] ||= []
         sortable_milestones[milestone[:milestone]] << timestamp
@@ -184,26 +183,26 @@ module Dor
 
       sortable_milestones.each do |milestone, unordered_dates|
         dates = unordered_dates.sort
-        #create the published_dttsi and published_day fields and the like
+        # create the published_dttsi and published_day fields and the like
         dates.each do |date|
           solr_doc["#{milestone}_dttsim"] ||= []
           solr_doc["#{milestone}_dttsim"] << date unless solr_doc["#{milestone}_dttsim"].include?(date)
         end
-        #fields for OAI havester to sort on: _dttsi is trie date +stored +indexed (single valued, i.e. sortable)
+        # fields for OAI havester to sort on: _dttsi is trie date +stored +indexed (single valued, i.e. sortable)
         solr_doc["#{milestone}_earliest_dttsi"] = dates.first
         solr_doc["#{milestone}_latest_dttsi"  ] = dates.last
       end
-      solr_doc["status_ssi"] = status # status is singular (i.e. the current one)
-      solr_doc["current_version_isi"] = current_version.to_i
-      solr_doc["modified_latest_dttsi"] = self.modified_date.to_datetime.utc.strftime('%FT%TZ')
-      add_solr_value(solr_doc, "rights", rights, :string, [:symbol]) if self.respond_to? :rights
+      solr_doc['status_ssi'] = status # status is singular (i.e. the current one)
+      solr_doc['current_version_isi'] = current_version.to_i
+      solr_doc['modified_latest_dttsi'] = modified_date.to_datetime.utc.strftime('%FT%TZ')
+      add_solr_value(solr_doc, 'rights', rights, :string, [:symbol]) if self.respond_to? :rights
 
-      status_info_hash = status_info()
+      status_info_hash = status_info
       status_code = status_info_hash[:status_code]
       add_solr_value(solr_doc, 'processing_status_text', simplified_status_code_disp_txt(status_code), :string, [:stored_sortable])
       solr_doc['processing_status_code_isi'] = status_code # no _isi in Solrizer's default descriptors
 
-      return solr_doc
+      solr_doc
     end
 
     # Initilizes workflow for the object in the workflow service
@@ -212,26 +211,30 @@ module Dor
     # @param [String] name of the workflow to be initialized
     # @param [Boolean] create_ds create a 'workflows' datastream in Fedora for the object
     # @param [Integer] priority the workflow's priority level
-    def initialize_workflow(name, create_ds=true, priority=0)
+    def create_workflow(name, create_ds = true, priority = 0)
       priority = workflows.current_priority if priority == 0
       opts = { :create_ds => create_ds, :lane_id => default_workflow_lane }
       opts[:priority] = priority if priority > 0
-      Dor::WorkflowService.create_workflow(Dor::WorkflowObject.initial_repo(name), self.pid, name, Dor::WorkflowObject.initial_workflow(name), opts)
+      Dor::WorkflowService.create_workflow(Dor::WorkflowObject.initial_repo(name), pid, name, Dor::WorkflowObject.initial_workflow(name), opts)
+      workflows.content(true) # refresh the copy of the workflows datastream
     end
 
-    private
-    #handles formating utc date/time to human readable
-    # XXX: bad form to hardcode TZ here.  Code smell abounds.
-    def format_date datetime
+    def initialize_workflow(name, create_ds = true, priority = 0)
+      warn 'WARNING: initialize_workflow is deprecated, use create_workflow instead'
+      create_workflow(name, create_ds, priority)
+    end
 
-        d = datetime.is_a?(Time) ? datetime :
-            DateTime.parse(datetime).in_time_zone(ActiveSupport::TimeZone.new("Pacific Time (US & Canada)"))
-        I18n.l(d).strftime('%Y-%m-%d %I:%M%p')
-      rescue
-        d = datetime.is_a?(Time) ? datetime : Time.parse(datetime.to_s)
-        d.strftime('%Y-%m-%d %I:%M%p')
+    private
 
+    # handles formating utc date/time to human readable
+    # XXX: bad form to hardcode TZ here.  Code smell abounds.
+    def format_date(datetime)
+      d = datetime.is_a?(Time) ? datetime :
+        DateTime.parse(datetime).in_time_zone(ActiveSupport::TimeZone.new('Pacific Time (US & Canada)'))
+      I18n.l(d).strftime('%Y-%m-%d %I:%M%p')
+    rescue
+      d = datetime.is_a?(Time) ? datetime : Time.parse(datetime.to_s)
+      d.strftime('%Y-%m-%d %I:%M%p')
     end
   end
-
 end

data/lib/dor/models/publishable.rb

@@ -7,14 +7,13 @@ module Dor
     include Governable
     include Describable
     include Itemizable
-    include Presentable
     include Rightsable
 
     def public_relationships
-      include_elements = ['fedora:isMemberOf','fedora:isMemberOfCollection']
-      rels_doc = Nokogiri::XML(self.datastreams['RELS-EXT'].content)
+      include_elements = ['fedora:isMemberOf', 'fedora:isMemberOfCollection', 'fedora:isConstituentOf']
+      rels_doc = Nokogiri::XML(datastreams['RELS-EXT'].content)
       rels_doc.xpath('/rdf:RDF/rdf:Description/*', { 'rdf' => 'http://www.w3.org/1999/02/22-rdf-syntax-ns#' }).each do |rel|
-        unless include_elements.include?([rel.namespace.prefix,rel.name].join(':'))
+        unless include_elements.include?([rel.namespace.prefix, rel.name].join(':'))
           rel.next_sibling.remove if rel.next_sibling.content.strip.empty?
           rel.remove
         end
@@ -22,29 +21,30 @@ module Dor
       rels_doc
     end
 
-    #Generate the public .xml for a PURL page.
-    #@return [xml] The public xml for the item
-    #
+    # Generate the public .xml for a PURL page.
+    # @return [xml] The public xml for the item
     def public_xml
-      pub = Nokogiri::XML("<publicObject/>").root
+      pub = Nokogiri::XML('<publicObject/>').root
       pub['id'] = pid
-      pub['published'] = Time.now.xmlschema
-      release_xml=Nokogiri(self.generate_release_xml).xpath('//release')
+      pub['published'] = Time.now.utc.xmlschema
+      pub['publishVersion'] = 'dor-services/' + Dor::VERSION
+      release_xml = Nokogiri(generate_release_xml).xpath('//release')
 
-      im=self.datastreams['identityMetadata'].ng_xml.clone
-      im.search('//release').each {|node| node.remove} # remove any <release> tags from public xml which have full history
+      im = datastreams['identityMetadata'].ng_xml.clone
+      im.search('//release').each(&:remove) # remove any <release> tags from public xml which have full history
       im.root.add_child(release_xml)
 
       pub.add_child(im.root) # add in modified identityMetadata datastream
-      pub.add_child(self.datastreams['contentMetadata'].public_xml.root.clone)
-      pub.add_child(self.datastreams['rightsMetadata'].ng_xml.root.clone)
+      pub.add_child(datastreams['contentMetadata'].public_xml.root.clone)
+      pub.add_child(datastreams['rightsMetadata'].ng_xml.root.clone)
 
       rels = public_relationships.root
       pub.add_child(rels.clone) unless rels.nil? # TODO: Should never be nil in practice; working around an ActiveFedora quirk for testing
-      pub.add_child(self.generate_dublin_core.root.clone)
+      pub.add_child(generate_dublin_core.root.clone)
       @public_xml_doc = pub # save this for possible IIIF Presentation manifest
-      pub.add_child(Nokogiri(self.generate_release_xml).root.clone) unless release_xml.children.size == 0 #If there are no release_tags, this prevents an empty <releaseData/> from being added
-      #Note we cannot base this on if an individual object has release tags or not, because the collection may cause one to be generated for an item, so we need to calculate it and then look at the final result
+      pub.add_child(Nokogiri(generate_release_xml).root.clone) unless release_xml.children.size == 0 # If there are no release_tags, this prevents an empty <releaseData/> from being added
+      # Note we cannot base this on if an individual object has release tags or not, because the collection may cause one to be generated for an item,
+      # so we need to calculate it and then look at the final result.s
       new_pub = Nokogiri::XML(pub.to_xml) { |x| x.noblanks }
       new_pub.encoding = 'UTF-8'
       new_pub.to_xml
@@ -55,29 +55,24 @@ module Dor
     def publish_metadata
       rights = datastreams['rightsMetadata'].ng_xml.clone.remove_namespaces!
       if rights.at_xpath("//rightsMetadata/access[@type='discover']/machine/world")
-        dc_xml = self.generate_dublin_core.to_xml {|config| config.no_declaration}
+        dc_xml = generate_dublin_core.to_xml {|config| config.no_declaration}
         DigitalStacksService.transfer_to_document_store(pid, dc_xml, 'dc')
-        DigitalStacksService.transfer_to_document_store(pid, self.datastreams['identityMetadata'].to_xml, 'identityMetadata')
-        DigitalStacksService.transfer_to_document_store(pid, self.datastreams['contentMetadata'].to_xml, 'contentMetadata')
-        DigitalStacksService.transfer_to_document_store(pid, self.datastreams['rightsMetadata'].to_xml, 'rightsMetadata')
-        DigitalStacksService.transfer_to_document_store(pid, public_xml, 'public')
-        if self.metadata_format == 'mods'
-          DigitalStacksService.transfer_to_document_store(pid, self.generate_public_desc_md, 'mods')
-        end
-        if iiif_presentation_manifest_needed? @public_xml_doc
-          DigitalStacksService.transfer_to_document_store(pid, build_iiif_manifest(@public_xml_doc), 'manifest')
+        %w(identityMetadata contentMetadata rightsMetadata).each do |stream|
+          DigitalStacksService.transfer_to_document_store(pid, datastreams[stream].to_xml, stream)
         end
+        DigitalStacksService.transfer_to_document_store(pid, public_xml, 'public')
+        DigitalStacksService.transfer_to_document_store(pid, generate_public_desc_md, 'mods') if metadata_format == 'mods'
       else
         # Clear out the document cache for this item
         DigitalStacksService.prune_purl_dir pid
       end
     end
-    #call the dor services app to have it publish the metadata
+
+    # Call dor services app to have it publish the metadata
     def publish_metadata_remotely
-      dor_services = RestClient::Resource.new(Config.dor_services.url+"/v1/objects/#{pid}/publish")
+      dor_services = RestClient::Resource.new(Config.dor_services.url + "/v1/objects/#{pid}/publish")
       dor_services.post ''
       dor_services.url
     end
   end
-
 end

data/lib/dor/models/releaseable.rb

@@ -11,18 +11,16 @@ module Dor
     # @param release_tags [Hash, Array<Hash>] hash of a single release tag or an array of many such hashes
     # @raise [ArgumentError] Raised if the tags are improperly supplied
     def add_release_nodes_and_start_releaseWF(release_tags)
-      release_tags = [release_tags] unless release_tags.class == Array
+      release_tags = [release_tags] unless release_tags.is_a?(Array)
 
-      #Add in each tag
+      # Add in each tag
       release_tags.each do |r_tag|
-        self.add_release_node(r_tag[:release],r_tag)
+        add_release_node(r_tag[:release], r_tag)
       end
 
-      #Save the item to dor so the robots work with the latest data
-      self.save
-
-      #Intialize the release workflow
-      self.initialize_workflow('releaseWF')
+      # Save item to dor so the robots work with the latest data
+      save
+      initialize_workflow('releaseWF')
     end
 
     # Generate XML structure for inclusion to Purl
@@ -30,70 +28,60 @@ module Dor
     def generate_release_xml
       builder = Nokogiri::XML::Builder.new do |xml|
         xml.releaseData {
-          self.released_for.each do |project, released_value|
-            xml.release(released_value["release"],:to=>project)
+          released_for.each do |project, released_value|
+            xml.release(released_value['release'], :to => project)
           end
         }
-        end
-      return builder.to_xml
+      end
+      builder.to_xml
     end
 
-    # Determine which projects an item is released for
+    # Determine projects in which an item is released
+    # @param [Boolean] skip_live_purl set true to skip requesting from purl backend
     # @return [Hash{String => Boolean}] all namespaces, keys are Project name Strings, values are Boolean
-    def released_for
+    def released_for(skip_live_purl = false)
       released_hash = {}
 
-      #Get release tags on the item itself
-      release_tags_on_this_item = self.release_nodes
-
-      #Get any self tags on this item
-      self_release_tags = self.get_self_release_tags(release_tags_on_this_item)
-
-      #Get the most recent self tag for all targets and save their result since most recent self always trumps any other non self tags
-      latest_self_tags = self.get_newest_release_tag(self_release_tags)
-      latest_self_tags.keys.each do |target|
-        released_hash[target] =  self.clean_release_tag_for_purl(latest_self_tags[target])
+      # Get the most recent self tag for all targets and retain their result since most recent self always trumps any other non self tags
+      latest_self_tags = get_newest_release_tag get_self_release_tags(release_nodes)
+      latest_self_tags.each do |key, payload|
+        released_hash[key] = {'release' => payload['release']}
       end
 
-      #With Self Tags Resolved We Now need to deal with tags on all sets this object is part of
-
-      potential_applicable_release_tags = {}  #This will be where we store all tags that apply, regardless of their timestamp
-
-      #Get all release tags on the item and strip out the what = self ones, we've already processed all the self tags on this item
-      potential_applicable_release_tags = get_tags_for_what_value(self.get_release_tags_for_item_and_all_governing_sets, 'collection')
-
-      administrative_tags = self.tags  #Get them once here and pass them down
-
-      #We now have the keys for all potential releases, we need to check the tags and the most recent time stamp with an explicit true or false wins, in a nil case, the lack of an explicit false tag we do nothing
-      (potential_applicable_release_tags.keys-released_hash.keys).each do |key|  #don't bother checking the ones already added to the release hash, they were added due to a self tag and that has won
-        latest_applicable_tag_for_key = latest_applicable_release_tag_in_array(potential_applicable_release_tags[key], administrative_tags)
-        unless latest_applicable_tag_for_key.nil? #We have a valid tag, record it
-          released_hash[key] = self.clean_release_tag_for_purl(latest_applicable_tag_for_key)
-        end
-
+      # With Self Tags resolved we now need to deal with tags on all sets this object is part of.
+      # Get all release tags on the item and strip out the what = self ones, we've already processed all the self tags on this item.
+      # This will be where we store all tags that apply, regardless of their timestamp:
+      potential_applicable_release_tags = get_tags_for_what_value(get_release_tags_for_item_and_all_governing_sets, 'collection')
+      administrative_tags = tags  # Get admin tags once here and pass them down
+
+      # We now have the keys for all potential releases, we need to check the tags: the most recent timestamp with an explicit true or false wins.
+      # In a nil case, the lack of an explicit false tag we do nothing.
+      (potential_applicable_release_tags.keys - released_hash.keys).each do |key|  # don't bother checking if already added to the release hash, they were added due to a self tag so that has won
+        latest_tag = latest_applicable_release_tag_in_array(potential_applicable_release_tags[key], administrative_tags)
+        next if latest_tag.nil? # Otherwise, we have a valid tag, record it
+        released_hash[key] = {'release' => latest_tag['release']}
       end
 
-      #See what the application is currently released for on Purl.  If something is released in purl but not listed here, it needs to be added as a false
-      released_hash = self.add_tags_from_purl(released_hash)
-
-      return released_hash
+      # See what the application is currently released for on Purl.  If released in purl but not listed here, it needs to be added as a false
+      add_tags_from_purl(released_hash) unless skip_live_purl
+      released_hash
     end
 
     # Take a hash of tags as obtained via Dor::Item.release_tags and returns all self tags
     # @param tags [Hash] a hash of tags obtained via Dor::Item.release_tags or matching format
     # @return [Hash] a hash of self tags for each to value
     def get_self_release_tags(tags)
-      return get_tags_for_what_value(tags, 'self')
+      get_tags_for_what_value(tags, 'self')
     end
 
     # Take an item and get all of its release tags and all tags on collections it is a member of it
     # @return [Hash] a hash of all tags
     def get_release_tags_for_item_and_all_governing_sets
-      return_tags = self.release_nodes || {}
-      self.collections.each do |collection|
-        return_tags = combine_two_release_tag_hashes(return_tags, Dor::Item.find(collection.id).get_release_tags_for_item_and_all_governing_sets) #this will function recurvisely so parents of parents are found
+      return_tags = release_nodes || {}
+      collections.each do |collection|
+        return_tags = combine_two_release_tag_hashes(return_tags, Dor::Item.find(collection.id).get_release_tags_for_item_and_all_governing_sets) # recurvise so parents of parents are found
       end
-      return return_tags
+      return_tags
     end
 
     # Take two hashes of tags and combine them, will not overwrite but will enforce uniqueness of the tags
@@ -105,7 +93,7 @@ module Dor
         hash_one[key] = hash_two[key] if hash_one[key].nil?
         hash_one[key] = (hash_one[key] + hash_two[key]).uniq unless hash_one[key].nil?
       end
-      return hash_one
+      hash_one
     end
 
     # Take a hash of tags and return all tags with the matching what target
@@ -115,34 +103,17 @@ module Dor
     def get_tags_for_what_value(tags, what_target)
       return_hash = {}
       tags.keys.each do |key|
-        self_tags =  tags[key].select{|tag| tag['what'] == what_target.downcase}
+        self_tags = tags[key].select {|tag| tag['what'] == what_target.downcase}
         return_hash[key] = self_tags if self_tags.size > 0
       end
-      return return_hash
+      return_hash
     end
 
     # Take a hash of tags as obtained via Dor::Item.release_tags and returns the newest tag for each namespace
     # @param tags [Hash] a hash of tags obtained via Dor::Item.release_tags or matching format
     # @return [Hash] a hash of latest tags for each to value
     def get_newest_release_tag(tags)
-      return_hash = {}
-      tags.keys.each do |key|
-        latest_for_key = newest_release_tag_in_an_array(tags[key])
-        return_hash[key] = latest_for_key
-      end
-      return return_hash
-    end
-
-    # Take a tag and return only the attributes  we want to put into purl
-    # @param tag [Hash] a tag
-    # @return [Hash] a hash of the attributes we want for purl
-    def clean_release_tag_for_purl(tag)
-      for_purl = ['release']
-      return_hash = {}
-      for_purl.each do |attr|
-        return_hash[attr] = tag[attr]
-      end
-      return return_hash
+      Hash[tags.map {|key, val| [key, newest_release_tag_in_an_array(val)]}]
     end
 
     # Takes an array of release tags and returns the most recent one
@@ -153,51 +124,50 @@ module Dor
       array_of_tags.each do |tag|
         latest_tag_in_array = tag if tag['when'] > latest_tag_in_array['when']
       end
-      return latest_tag_in_array
+      latest_tag_in_array
     end
 
     # Takes a tag and returns true or false if it applies to the specific item
     # @param release_tag [Hash] the tag in a hashed form
     # @param admin_tags [Array] the administrative tags on an item, if not supplied it will attempt to retrieve them
     # @return [Boolean] true or false if it applies (not true or false if it is released, that is the release_tag data)
-    def does_release_tag_apply(release_tag, admin_tags=false)
-      #Is the tag global or restricted
-      return true if release_tag['tag'].nil?  #there is no specific tag specificied, so that means this tag is global to all members of the collection, it applies, return true
-
-      admin_tags = self.tags unless admin_tags #We use false instead of [], since an item can have no admin_tags that which point we'd be passing down this variable as [] and would not an attempt to retrieve it
-      return admin_tags.include?(release_tag['tag'])
+    def does_release_tag_apply(release_tag, admin_tags = false)
+      # Is the tag global or restricted
+      return true if release_tag['tag'].nil?  # no specific tag specificied means this tag is global to all members of the collection
+      admin_tags = tags unless admin_tags     # We use false instead of [], since an item can have no admin_tags at which point we'd be passing this var as [] and would not attempt to retrieve it
+      admin_tags.include?(release_tag['tag'])
     end
 
     # Takes an array of release tags and returns the most recent one that applies to this item
     # @param release_tags [Array] an array of release tags in hashed form
     # @param admin_tags [Array] the administrative tags on an on item
-    # @return [Hash] the tag
+    # @return [Hash] the tag, or nil if none applicable
     def latest_applicable_release_tag_in_array(release_tags, admin_tags)
       newest_tag = newest_release_tag_in_an_array(release_tags)
-      return newest_tag if does_release_tag_apply(newest_tag, admin_tags) #Return true if we have it
+      return newest_tag if does_release_tag_apply(newest_tag, admin_tags)
 
-      #The latest tag wasn't applicable, slice it off and try again
-      #This could be optimized by reordering on the timestamp and just running down it instead of constantly resorting, at least if we end up getting numerous release tags on an item
+      # The latest tag wasn't applicable, slice it off and try again
+      # This could be optimized by reordering on the timestamp and just running down it instead of constantly resorting, at least if we end up getting numerous release tags on an item
       release_tags.slice!(release_tags.index(newest_tag))
 
-      return latest_applicable_release_tag_in_array(release_tags, admin_tags) if release_tags.size > 0 #Try again after dropping the one that wasn't applicable
-      return nil #We're out of tags, no applicable ones
+      return latest_applicable_release_tag_in_array(release_tags, admin_tags) if release_tags.size > 0 # Try again after dropping the inapplicable
+      nil # We're out of tags, no applicable ones
     end
 
     # Helper method to get the release tags as a nodeset
     # @return [Nokogiri::XML::NodeSet] all release tags and their attributes
     def release_tags
-      release_tags = self.identityMetadata.ng_xml.xpath('//release')
+      release_tags = identityMetadata.ng_xml.xpath('//release')
       return_hash = {}
       release_tags.each do |release_tag|
-        hashed_node = self.release_tag_node_to_hash(release_tag)
+        hashed_node = release_tag_node_to_hash(release_tag)
         if !return_hash[hashed_node[:to]].nil?
           return_hash[hashed_node[:to]] << hashed_node[:attrs]
         else
           return_hash[hashed_node[:to]] = [hashed_node[:attrs]]
         end
       end
-      return return_hash
+      return_hash
     end
 
     # Convert one release element into a Hash
@@ -206,21 +176,20 @@ module Dor
     def release_tag_node_to_hash(rtag)
       to = 'to'
       release = 'release'
-      when_word = 'when' #TODO: Make to and when_word load from some config file instead of hardcoded here
+      when_word = 'when' # TODO: Make to and when_word load from some config file instead of hardcoded here
       attrs = rtag.attributes
       return_hash = { :to => attrs[to].value }
-      attrs.tap { |a| a.delete(to)}
-      attrs[release] = rtag.text.downcase == "true" #save release as a boolean
+      attrs.tap { |a| a.delete(to) }
+      attrs[release] = rtag.text.downcase == 'true' # save release as a boolean
       return_hash[:attrs] = attrs
 
-      #convert all the attrs beside :to to strings, they are currently Nokogiri::XML::Attr
-      (return_hash[:attrs].keys-[to]).each do |a|
-        return_hash[:attrs][a] =  return_hash[:attrs][a].to_s if a != release
+      # convert all the attrs beside :to to strings, they are currently Nokogiri::XML::Attr
+      (return_hash[:attrs].keys - [to]).each do |a|
+        return_hash[:attrs][a] = return_hash[:attrs][a].to_s if a != release
       end
 
-      return_hash[:attrs][when_word] = Time.parse(return_hash[:attrs][when_word]) #convert when to a datetime
-
-      return return_hash
+      return_hash[:attrs][when_word] = Time.parse(return_hash[:attrs][when_word]) # convert when to a datetime
+      return_hash
     end
 
     # Determine if the supplied tag is a valid release tag that meets all requirements
@@ -228,8 +197,8 @@ module Dor
     # @param attrs [hash] A hash of attributes for the tag, must contain: :when, a ISO 8601 timestamp; :who, to identify who or what added the tag; and :to, a string identifying the release target
     # @raise [RuntimeError]  Raises an error of the first fault in the release tag
     # @return [Boolean] Returns true if no errors found
-    def valid_release_attributes_and_tag(tag, attrs={})
-      raise ArgumentError, ":when is not iso8601" if attrs[:when].match('\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}Z').nil?
+    def valid_release_attributes_and_tag(tag, attrs = {})
+      raise ArgumentError, ':when is not iso8601' if attrs[:when].match('\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}Z').nil?
       [:who, :to, :what].each do |check_attr|
         raise ArgumentError, "#{check_attr} not supplied as a String" if attrs[check_attr].class != String
       end
@@ -238,10 +207,10 @@ module Dor
       %w(self collection).each do |allowed_what_value|
         what_correct = true if attrs[:what] == allowed_what_value
       end
-      raise ArgumentError, ":what must be self or collection" unless what_correct
-      raise ArgumentError, "the value set for this tag is not a boolean" if !!tag != tag
-      validate_tag_format(attrs[:tag]) unless attrs[:tag].nil? #Will Raise exception if invalid tag
-      return true
+      raise ArgumentError, ':what must be self or collection' unless what_correct
+      raise ArgumentError, 'the value set for this tag is not a boolean' if !!tag != tag # rubocop:disable Style/DoubleNegation
+      validate_tag_format(attrs[:tag]) unless attrs[:tag].nil? # Will Raise exception if invalid tag
+      true
     end
 
     # Add a release node for the item
@@ -255,17 +224,16 @@ module Dor
     #
     # @example
     #  item.add_tag(true,:release,{:tag=>'Fitch : Batch2',:what=>'self',:to=>'Searchworks',:who=>'petucket', :displayType='filmstrip'})
-    def add_release_node(release, attrs={})
-      identity_metadata_ds = self.identityMetadata
-      attrs[:when] = Time.now.utc.iso8601 if attrs[:when].nil? #add the timestamp
-      attrs[:displayType] = 'file' if attrs[:displayType].nil? #default to file is no display type is passed
+    def add_release_node(release, attrs = {})
+      identity_metadata_ds = identityMetadata
+      attrs[:when] = Time.now.utc.iso8601 if attrs[:when].nil? # add the timestamp
+      attrs[:displayType] = 'file' if attrs[:displayType].nil? # default to file is no display type is passed
       valid_release_attributes(release, attrs)
 
-      #Remove the old displayType and then add the one for this tag
+      # Remove the old displayType and then add the one for this tag
       remove_displayTypes
       identity_metadata_ds.add_value(:displayType, attrs[:displayType], {})
-
-      return identity_metadata_ds.add_value(:release, release.to_s, attrs)
+      identity_metadata_ds.add_value(:release, release.to_s, attrs)
     end
 
     # Determine if the supplied tag is a valid release node that meets all requirements
@@ -274,8 +242,8 @@ module Dor
     # @param attrs [hash] A hash of attributes for the tag, must contain :when, a ISO 8601 timestamp and :who to identify who or what added the tag, :to,
     # @raise [ArgumentError]  Raises an error of the first fault in the release tag
     # @return [Boolean] Returns true if no errors found
-    def valid_release_attributes(tag, attrs={})
-      raise ArgumentError, ":when is not iso8601" if attrs[:when].match('\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}Z').nil?
+    def valid_release_attributes(tag, attrs = {})
+      raise ArgumentError, ':when is not iso8601' if attrs[:when].match('\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}Z').nil?
       [:who, :to, :what].each do |check_attr|
         raise ArgumentError, "#{check_attr} not supplied as a String" if attrs[check_attr].class != String
       end
@@ -284,28 +252,28 @@ module Dor
       %w(self collection).each do |allowed_what_value|
         what_correct = true if attrs[:what] == allowed_what_value
       end
-      raise ArgumentError, ":what must be self or collection" unless what_correct
-      raise ArgumentError, "the value set for this tag is not a boolean" if !!tag != tag
-      raise ArgumentError, ":displayType must be passed in as a String" unless attrs[:displayType].class == String
+      raise ArgumentError, ':what must be self or collection' unless what_correct
+      raise ArgumentError, 'the value set for this tag is not a boolean' if !!tag != tag # rubocop:disable Style/DoubleNegation
+      raise ArgumentError, ':displayType must be passed in as a String' unless attrs[:displayType].class == String
 
-      validate_tag_format(attrs[:tag]) unless attrs[:tag].nil? #Will Raise exception if invalid tag
-      return true
+      validate_tag_format(attrs[:tag]) unless attrs[:tag].nil? # Will Raise exception if invalid tag
+      true
     end
 
     # Helper method to get the release nodes as a nodeset
     # @return [Nokogiri::XML::NodeSet] of all release tags and their attributes
     def release_nodes
-      release_tags = self.identityMetadata.ng_xml.xpath('//release')
+      release_tags = identityMetadata.ng_xml.xpath('//release')
       return_hash = {}
       release_tags.each do |release_tag|
-        hashed_node = self.release_tag_node_to_hash(release_tag)
+        hashed_node = release_tag_node_to_hash(release_tag)
         if !return_hash[hashed_node[:to]].nil?
           return_hash[hashed_node[:to]] << hashed_node[:attrs]
         else
-           return_hash[hashed_node[:to]] = [hashed_node[:attrs]]
+          return_hash[hashed_node[:to]] = [hashed_node[:attrs]]
         end
       end
-      return return_hash
+      return_hash
     end
 
     # Get a list of all release nodes found in a purl document
@@ -313,81 +281,76 @@ module Dor
     # @raise [OpenURI::HTTPError]
     # @return [Nokogiri::HTML::Document] parsed XML for the druid or an empty document if no purl is found
     def get_xml_from_purl
-      handler = Proc.new do |exception, attempt_number, total_delay|
-        #We assume a 404 means the document has never been published before and thus has no purl
-        #The strip is needed before the actual message is "404 "
-        return Nokogiri::HTML::Document.new if exception.message.strip == "404"
+      url = form_purl_url
+      handler = proc do |exception, attempt_number, total_delay|
+        # We assume a 404 means the document has never been published before and thus has no purl
+        Dor.logger.warn "[Attempt #{attempt_number}] GET #{url} -- #{exception.class}: #{exception.message}; #{total_delay} seconds elapsed."
+        raise exception unless exception.is_a? OpenURI::HTTPError
+        return Nokogiri::HTML::Document.new if exception.message.strip == '404'    # strip is needed if the actual message is "404 "
       end
 
-      with_retries(:max_retries => 5, :base_sleep_seconds => 3, :max_sleep_seconds=> 5, :rescue => OpenURI::HTTPError, :handler => handler) {
-         #If you change the method used for opening the webpage, you can change the :rescue param to handle the new method's errors
-         return Nokogiri::HTML(open(self.form_purl_url))
-       }
+      with_retries(:max_retries => 3, :base_sleep_seconds => 3, :max_sleep_seconds => 5, :handler => handler) do |attempt|
+        # If you change the method used for opening the webpage, you can change the :rescue param to handle the new method's errors
+        Dor.logger.info "[Attempt #{attempt}] GET #{url}"
+        return Nokogiri::HTML(OpenURI.open_uri(url))
+      end
     end
 
     # Since purl does not use the druid: prefix but much of dor does, use this function to strip the druid: if needed
     # @return [String] the druid sans the druid: or if there was no druid: prefix, the entire string you passed
     def remove_druid_prefix
-      druid_prefix = "druid:"
-      return self.id.split(druid_prefix)[1] if self.id.split(druid_prefix).size > 1
-      return druid
+      druid_prefix = 'druid:'
+      return id.split(druid_prefix)[1] if id.split(druid_prefix).size > 1
+      druid
     end
 
     # Take the and create the entire purl url that will usable for the open method in open-uri, returns http
-    # @return [String], the full url
+    # @return [String] the full url
     def form_purl_url
-      prefix = "http://"
-      return prefix + Dor::Config.stacks.document_cache_host + "/#{self.remove_druid_prefix}.xml"
+      'https://' + Dor::Config.stacks.document_cache_host + "/#{remove_druid_prefix}.xml"
     end
 
     # Pull all release nodes from the public xml obtained via the purl query
     # @param doc [Nokogiri::HTML::Document] The druid of the object you want
     # @return [Array] An array containing all the release tags
     def get_release_tags_from_purl_xml(doc)
-      nodes = doc.xpath("//html/body/publicobject/releasedata").children
-      #We only want the nodes with a name that isn't text
-      return_array = []
-      nodes.each do |n|
-        return_array << n.attr('to') if !n.name.nil? && n.name.downcase != "text"
-      end
-      return return_array.uniq
+      nodes = doc.xpath('//html/body/publicobject/releasedata').children
+      # We only want the nodes with a name that isn't text
+      nodes.reject {|n| n.name.nil? || n.name.downcase == 'text'}.map {|n| n.attr('to')}.uniq
     end
 
     # Pull all release nodes from the public xml obtained via the purl query
     # @return [Array] An array containing all the release tags
     def get_release_tags_from_purl
-      xml = self.get_xml_from_purl
-      return self.get_release_tags_from_purl_xml(xml)
+      get_release_tags_from_purl_xml(get_xml_from_purl)
     end
 
     # This function calls purl and gets a list of all release tags currently in purl.  It then compares to the list you have generated.
     # Any tag that is on purl, but not in the newly generated list is added to the new list with a value of false.
-    #
     # @param new_tags [Hash{String => Boolean}] all new tags in the form of !{"Project" => Boolean}
-    # @return [Hash], a hash in the same form as new_tags, with all missing tags not in new_tags, but in current_tag_names, added in with a Boolean value of false
+    # @return [Hash] same form as new_tags, with all missing tags not in new_tags, but in current_tag_names, added in with a Boolean value of false
     def add_tags_from_purl(new_tags)
-      tags_currently_in_purl = self.get_release_tags_from_purl
+      tags_currently_in_purl = get_release_tags_from_purl
       missing_tags = tags_currently_in_purl.map(&:downcase) - new_tags.keys.map(&:downcase)
       missing_tags.each do |missing_tag|
-        new_tags[missing_tag.capitalize] = {"release"=>false}
+        new_tags[missing_tag.capitalize] = {'release' => false}
       end
-      return new_tags
+      new_tags
     end
 
-    def to_solr(solr_doc=Hash.new, *args)
+    def to_solr(solr_doc = {}, *args)
       super(solr_doc, *args)
 
-      #TODO: sort of worried about the performance impact in bulk reindex
+      # TODO: sort of worried about the performance impact in bulk reindex
       # situations, since released_for recurses all parent collections.  jmartin 2015-07-14
-      released_for().each { |key, val|
-        add_solr_value(solr_doc, "released_to", key, :symbol, []) if val
+      released_for(true).each { |key, val|
+        add_solr_value(solr_doc, 'released_to', key, :symbol, []) if val
       }
 
-      #TODO: need to solrize whether item is released to purl?  does released_for
-      # return that?  logic is:  "True when there is a published lifecycle and Access
-      # Rights is anything but Dark"
+      # TODO: need to solrize whether item is released to purl?  does released_for return that?
+      # logic is: "True when there is a published lifecycle and Access Rights is anything but Dark"
 
-      return solr_doc
+      solr_doc
     end
   end
 end