arclight 0.1.4 → 0.3.3

data/lib/arclight/engine.rb

@@ -1,33 +1,36 @@
 # frozen_string_literal: true
 
+require 'sprockets/bumble_d'
 require 'blacklight'
-require 'solr_ead'
+require 'traject'
+require 'active_model'
 require 'arclight/exceptions'
 require 'arclight/normalized_date'
 require 'arclight/normalized_id'
 require 'arclight/normalized_title'
 require 'arclight/digital_object'
-require 'arclight/shared_indexing_behavior'
-require 'arclight/shared_terminology_behavior'
-require 'arclight/custom_document'
-require 'arclight/custom_component'
-require 'arclight/solr_ead_indexer_ext'
-require 'arclight/indexer'
 require 'arclight/viewer'
 
 module Arclight
   ##
   # This is the defining class for the Arclight Rails Engine
   class Engine < ::Rails::Engine
+    extend ::Sprockets::BumbleD::DSL
+
+    # We're not sure this is right, but we aren't doing module imports
+    # at the moment anyway
+    register_umd_globals :arclight,
+                         'blacklight' => 'Blacklight'
+
     config.viewer_class = Arclight::Viewers::OEmbed
     config.oembed_resource_exclude_patterns = [/\.pdf$/, /\.ppt$/]
 
     Arclight::Engine.config.catalog_controller_field_accessors = %i[
       summary_field
       access_field
+      contact_field
       background_field
       related_field
-      admin_info_field
       terms_field
       cite_field
       indexed_terms_field
@@ -38,6 +41,18 @@ module Arclight
       component_indexed_terms_field
     ]
 
+    Arclight::Engine.config.catalog_controller_group_query_params = {
+      group: true,
+      'group.field': 'collection_ssi',
+      'group.ngroups': true,
+      'group.limit': 3,
+      fl: '*,parent:[subquery]',
+      'parent.fl': '*',
+      'parent.q': '{!term f=collection_sim v=$row.collection_ssi}',
+      'parent.fq': '{!term f=level_sim v="Collection"}',
+      'parent.defType': 'lucene'
+    }
+
     initializer 'arclight.fields' do
       Arclight::Engine.config.catalog_controller_field_accessors.each do |field|
         Blacklight::Configuration.define_field_access field
@@ -47,9 +62,5 @@ module Arclight
     initializer 'arclight.helpers' do
       ActionView::Base.send :include, ArclightHelper
     end
-
-    initializer 'arclight.views' do
-      Blacklight::Configuration.default_values[:view].hierarchy
-    end
   end
 end

data/lib/arclight/hash_absolute_xpath.rb

@@ -0,0 +1,61 @@
+# frozen_string_literal: true
+
+require 'digest'
+
+module Arclight
+  ##
+  # Take a Nokogiri node and get its absolute path (inserting our own indexes for component levels)
+  # and hash that outout. This is intended as a potential strategy for handling missing IDs in EADs.
+  class HashAbsoluteXpath
+    class << self
+      attr_writer :hash_algorithm
+
+      def hash_algorithm
+        return Digest::SHA1 unless defined? @hash_algorithm
+
+        @hash_algorithm
+      end
+    end
+
+    COMPONENT_NODE_NAME_REGEX = /^c\d{,2}$/.freeze
+    attr_reader :node
+    def initialize(node)
+      @node = node
+    end
+
+    def to_hexdigest
+      self.class.hash_algorithm.hexdigest(absolute_xpath).prepend('al_')
+    end
+
+    def absolute_xpath
+      ancestor_tree = node.ancestors.map do |ancestor|
+        ancestor_name_and_index(ancestor)
+      end
+
+      "#{[ancestor_tree.reverse, node.name].flatten.join('/')}#{current_index}"
+    end
+
+    private
+
+    def current_index
+      siblings.index(node)
+    end
+
+    def component_siblings_for_node(xml_node)
+      xml_node.parent.children.select { |n| n.name =~ COMPONENT_NODE_NAME_REGEX }
+    end
+
+    def siblings
+      @siblings ||= component_siblings_for_node(node)
+    end
+
+    def ancestor_name_and_index(ancestor)
+      if ancestor.name =~ COMPONENT_NODE_NAME_REGEX
+        index = component_siblings_for_node(ancestor).index(ancestor)
+        "#{ancestor.name}#{index}"
+      else
+        ancestor.name
+      end
+    end
+  end
+end

data/lib/arclight/level_label.rb

@@ -0,0 +1,46 @@
+# frozen_string_literal: true
+
+module Arclight
+  ##
+  # A utility class to return a human-readable label for an EAD @level code.
+  # Can use the value from @otherlevel if provided.
+  # Examples from @level: recordgrp = "Record Group"
+  #                       collection = "Collection"
+  #                       subseries = "Subseries"
+  #                       otherlevel = (text provided in @otherlevel)
+  class LevelLabel
+    # @param [String] `level` from the collection or component @level
+    # @param [String] `other_level` from the collection or component @otherlevel
+    def initialize(level, other_level = nil)
+      @level = level
+      @other_level = other_level if other_level.present?
+    end
+
+    # @return [String] the human-readable label
+    def to_s
+      human_readable_level
+    end
+
+    private
+
+    attr_reader :level, :other_level
+
+    CUSTOM_LEVEL_LABELS = {
+      recordgrp: 'Record Group',
+      subgrp: 'Subgroup'
+    }.freeze
+
+    def human_readable_level
+      if level == 'otherlevel'
+        alternative_level
+      elsif level.present?
+        CUSTOM_LEVEL_LABELS.fetch(level.to_sym, level.capitalize).to_s
+      end
+    end
+
+    def alternative_level
+      alternative_level = other_level if other_level
+      alternative_level.present? ? alternative_level.capitalize : 'Other'
+    end
+  end
+end

data/lib/arclight/missing_id_strategy.rb

@@ -0,0 +1,21 @@
+# frozen_string_literal: true
+
+require 'arclight/hash_absolute_xpath'
+
+module Arclight
+  ##
+  # A class to configure a selected MissingIdStrategy.
+  # Defaults to Arclight::HashAbsoluteXpath
+  # This can be updated in an initializer to be any other class
+  class MissingIdStrategy
+    class << self
+      attr_writer :selected
+
+      def selected
+        return Arclight::HashAbsoluteXpath unless defined? @selected
+
+        @selected
+      end
+    end
+  end
+end

data/lib/arclight/normalized_date.rb

@@ -7,16 +7,21 @@ module Arclight
   # @see http://www2.archivists.org/standards/DACS/part_I/chapter_2/4_date
   class NormalizedDate
     # @param [String | Array<String>] `inclusive` from the `unitdate`
-    # @param [String] `bulk` from the `unitdate`
-    # @param [String] `other` from the `unitdate` when type is not specified
-    def initialize(inclusive, bulk = nil, other = nil)
-      if inclusive.is_a? Array # of YYYY-YYYY for ranges
-        @inclusive = YearRange.new(inclusive.include?('/') ? inclusive : inclusive.map { |v| v.tr('-', '/') }).to_s
-      elsif inclusive.present?
-        @inclusive = inclusive.strip
-      end
-      @bulk = bulk.strip if bulk.present?
-      @other = other.strip if other.present?
+    # @param [Array<String>] `bulk` from the `unitdate`
+    # @param [Array<String>] `other` from the `unitdate` when type is not specified
+    def initialize(inclusive, bulk = [], other = [])
+      @inclusive = (inclusive || []).map do |inclusive_text|
+        if inclusive_text.is_a? Array # of YYYY-YYYY for ranges
+          # NOTE: This code is not routable AFAICT in actual indexing.
+          # We pass arrays of strings (or xml nodes) here, and never a multidimensional array
+          year_range(inclusive_text)
+        elsif inclusive_text.present?
+          inclusive_text.strip
+        end
+      end&.join(', ')
+
+      @bulk = Array.wrap(bulk).compact.map(&:strip).join(', ')
+      @other = Array.wrap(other).compact.map(&:strip).join(', ')
     end
 
     # @return [String] the normalized title/date
@@ -28,6 +33,10 @@ module Arclight
 
     attr_reader :inclusive, :bulk, :other
 
+    def year_range(date_array)
+      YearRange.new(date_array.include?('/') ? date_array : date_array.map { |v| v.tr('-', '/') }).to_s
+    end
+
     # @see http://www2.archivists.org/standards/DACS/part_I/chapter_2/4_date for rules
     def normalize
       if inclusive.present?
@@ -38,8 +47,8 @@ module Arclight
       else
         result = nil
       end
-      return if result.blank?
-      result.strip
+
+      result&.strip
     end
   end
 end

data/lib/arclight/normalized_id.rb

@@ -19,6 +19,7 @@ module Arclight
 
     def normalize
       raise Arclight::Exceptions::IDNotFound if id.blank?
+
       id.strip.tr('.', '-')
     end
   end

data/lib/arclight/normalized_title.rb

@@ -24,6 +24,7 @@ module Arclight
     def normalize
       result = [title, date].compact.join(', ')
       raise Arclight::Exceptions::TitleNotFound if result.blank?
+
       result
     end
   end

data/lib/arclight/repository.rb

@@ -7,31 +7,14 @@ module Arclight
   class Repository
     include ActiveModel::Conversion # for to_partial_path
 
-    FIELDS = %i[name
-                description
-                visit_note
-                building
-                address1
-                address2
-                city
-                state
-                zip
-                country
-                phone
-                contact_info
-                thumbnail_url
-                google_request_url
-                google_request_mappings
-                collection_count].freeze
-
-    attr_accessor :slug, *FIELDS
+    attr_accessor :slug, :collection_count
 
     # @param [String] `slug` the unique identifier for the repository
     # @param [Hash] `data`
     def initialize(slug, data = {})
       @slug = slug
-      FIELDS.each do |field|
-        value = data[field.to_s]
+      data.each do |field, value|
+        self.class.attr_accessor field.to_sym
         send("#{field}=", value) if value.present?
       end
     end
@@ -43,6 +26,57 @@ module Arclight
       [city, state_zip, country].compact.join(', ')
     end
 
+    # Why are we using self#respond_to? below?
+    #
+    # All the keys in the config hash from `repositories.yml` are
+    # on-the-fly added as attr_accessors up in #initialize. If the
+    # request_types key isn't present, the method won't be created.
+    #
+    # Since the original data is thrown away, this is the best way
+    # to see if that key was present.
+    def request_config_present?
+      return false unless respond_to? :request_types
+      return false if request_types.nil? || request_types.empty?
+
+      request_configs = request_types.map { |_k, v| v }
+      request_configs[0]&.fetch('request_url').present? &&
+        request_configs[0]&.fetch('request_mappings').present?
+    end
+
+    def request_config_present_for_type?(type)
+      return false unless type && request_config_present?
+
+      config = request_types[type]
+      config&.fetch('request_url').present? &&
+        config&.fetch('request_mappings').present?
+    end
+
+    def request_config_for_type(type)
+      return nil unless type && request_config_present_for_type?(type)
+
+      request_types[type]
+    end
+
+    def request_url_for_type(type)
+      return nil unless type && request_config_present_for_type?(type)
+
+      config = request_config_for_type(type)
+      config.fetch('request_url')
+    end
+
+    def request_mappings_for_type(type)
+      return nil unless type && request_config_present_for_type?(type)
+
+      config = request_config_for_type(type)
+      config.fetch('request_mappings')
+    end
+
+    def available_request_types
+      return [] unless request_types.present?
+
+      request_types.keys
+    end
+
     # Load repository information from a YAML file
     #
     # @param [String] `filename`
@@ -59,19 +93,20 @@ module Arclight
     # Mimics ActiveRecord's `all` behavior
     #
     # @return [Array<Repository>]
-    def self.all
-      from_yaml(ENV['REPOSITORY_FILE'] || 'config/repositories.yml').values
+    def self.all(yaml_file = nil)
+      yaml_file = ENV['REPOSITORY_FILE'] || 'config/repositories.yml' if yaml_file.nil?
+      from_yaml(yaml_file).values
     end
 
     # Mimics ActiveRecord dynamic `find_by` behavior for the slug or name
     #
     # @param [String] `slug` or `name`
     # @return [Repository]
-    def self.find_by(slug: nil, name: nil)
+    def self.find_by(slug: nil, name: nil, yaml_file: nil)
       if slug
-        all.find { |repo| repo.slug == slug }
+        all(yaml_file).find { |repo| repo.slug == slug }
       elsif name
-        all.find { |repo| repo.name == name }
+        all(yaml_file).find { |repo| repo.name == name }
       else
         raise ArgumentError, 'Requires either slug or name parameters to find_by'
       end
@@ -85,6 +120,7 @@ module Arclight
     def self.find_by!(*args)
       repository = find_by(*args)
       raise ActiveRecord::RecordNotFound if repository.blank?
+
       repository
     end
   end

data/lib/arclight/traject/ead2_config.rb

@@ -0,0 +1,495 @@
+# frozen_string_literal: true
+
+require 'logger'
+require 'traject'
+require 'traject/nokogiri_reader'
+require 'traject_plus'
+require 'traject_plus/macros'
+require 'arclight/exceptions'
+require 'arclight/level_label'
+require 'arclight/normalized_date'
+require 'arclight/normalized_title'
+require 'active_model/conversion' ## Needed for Arclight::Repository
+require 'active_support/core_ext/array/wrap'
+require 'arclight/digital_object'
+require 'arclight/year_range'
+require 'arclight/repository'
+require 'arclight/missing_id_strategy'
+require 'arclight/traject/nokogiri_namespaceless_reader'
+
+# rubocop:disable Style/MixinUsage
+extend TrajectPlus::Macros
+# rubocop:enable Style/MixinUsage
+
+NAME_ELEMENTS = %w[corpname famname name persname].freeze
+
+SEARCHABLE_NOTES_FIELDS = %w[
+  accessrestrict
+  accruals
+  altformavail
+  appraisal
+  arrangement
+  bibliography
+  bioghist
+  custodhist
+  fileplan
+  note
+  odd
+  originalsloc
+  otherfindaid
+  phystech
+  prefercite
+  processinfo
+  relatedmaterial
+  scopecontent
+  separatedmaterial
+  userestrict
+].freeze
+
+DID_SEARCHABLE_NOTES_FIELDS = %w[
+  abstract
+  materialspec
+  physloc
+].freeze
+
+settings do
+  provide 'reader_class_name', 'Arclight::Traject::NokogiriNamespacelessReader'
+  provide 'solr_writer.commit_on_close', 'true'
+  provide 'repository', ENV['REPOSITORY_ID']
+  provide 'logger', Logger.new($stderr)
+end
+
+each_record do |_record, context|
+  next unless settings['repository']
+
+  context.clipboard[:repository] = Arclight::Repository.find_by(
+    slug: settings['repository']
+  ).name
+end
+
+# ==================
+# Top level document
+# ==================
+
+to_field 'id', extract_xpath('/ead/eadheader/eadid'), strip, gsub('.', '-')
+to_field 'title_filing_si', extract_xpath('/ead/eadheader/filedesc/titlestmt/titleproper[@type="filing"]')
+to_field 'title_ssm', extract_xpath('/ead/archdesc/did/unittitle')
+to_field 'title_teim', extract_xpath('/ead/archdesc/did/unittitle')
+to_field 'ead_ssi', extract_xpath('/ead/eadheader/eadid')
+
+to_field 'unitdate_ssm', extract_xpath('/ead/archdesc/did/unitdate')
+to_field 'unitdate_bulk_ssim', extract_xpath('/ead/archdesc/did/unitdate[@type="bulk"]')
+to_field 'unitdate_inclusive_ssm', extract_xpath('/ead/archdesc/did/unitdate[@type="inclusive"]')
+to_field 'unitdate_other_ssim', extract_xpath('/ead/archdesc/did/unitdate[not(@type)]')
+
+# All top-level docs treated as 'collection' for routing / display purposes
+to_field 'level_ssm' do |_record, accumulator|
+  accumulator << 'collection'
+end
+
+# Keep the original top-level archdesc/@level for Level facet in addition to 'Collection'
+to_field 'level_sim' do |record, accumulator|
+  level = record.at_xpath('/ead/archdesc').attribute('level')&.value
+  other_level = record.at_xpath('/ead/archdesc').attribute('otherlevel')&.value
+
+  accumulator << Arclight::LevelLabel.new(level, other_level).to_s
+  accumulator << 'Collection' unless level == 'collection'
+end
+
+to_field 'unitid_ssm', extract_xpath('/ead/archdesc/did/unitid')
+to_field 'unitid_teim', extract_xpath('/ead/archdesc/did/unitid')
+to_field 'collection_unitid_ssm', extract_xpath('/ead/archdesc/did/unitid')
+
+to_field 'normalized_title_ssm' do |_record, accumulator, context|
+  dates = Arclight::NormalizedDate.new(
+    context.output_hash['unitdate_inclusive_ssm'],
+    context.output_hash['unitdate_bulk_ssim'],
+    context.output_hash['unitdate_other_ssim']
+  ).to_s
+  title = context.output_hash['title_ssm'].first
+  accumulator << Arclight::NormalizedTitle.new(title, dates).to_s
+end
+
+to_field 'normalized_date_ssm' do |_record, accumulator, context|
+  accumulator << Arclight::NormalizedDate.new(
+    context.output_hash['unitdate_inclusive_ssm'],
+    context.output_hash['unitdate_bulk_ssim'],
+    context.output_hash['unitdate_other_ssim']
+  ).to_s
+end
+
+to_field 'collection_ssm' do |_record, accumulator, context|
+  accumulator.concat context.output_hash.fetch('normalized_title_ssm', [])
+end
+to_field 'collection_sim' do |_record, accumulator, context|
+  accumulator.concat context.output_hash.fetch('normalized_title_ssm', [])
+end
+to_field 'collection_ssi' do |_record, accumulator, context|
+  accumulator.concat context.output_hash.fetch('normalized_title_ssm', [])
+end
+to_field 'collection_title_tesim' do |_record, accumulator, context|
+  accumulator.concat context.output_hash.fetch('normalized_title_ssm', [])
+end
+
+to_field 'repository_ssm' do |_record, accumulator, context|
+  accumulator << context.clipboard[:repository]
+end
+
+to_field 'repository_sim' do |_record, accumulator, context|
+  accumulator << context.clipboard[:repository]
+end
+
+to_field 'geogname_ssm', extract_xpath('/ead/archdesc/controlaccess/geogname')
+to_field 'geogname_sim', extract_xpath('/ead/archdesc/controlaccess/geogname')
+
+to_field 'creator_ssm', extract_xpath('/ead/archdesc/did/origination')
+to_field 'creator_sim', extract_xpath('/ead/archdesc/did/origination')
+to_field 'creator_ssim', extract_xpath('/ead/archdesc/did/origination')
+to_field 'creator_sort' do |record, accumulator|
+  accumulator << record.xpath('/ead/archdesc/did/origination').map { |c| c.text.strip }.join(', ')
+end
+
+to_field 'creator_persname_ssm', extract_xpath('/ead/archdesc/did/origination/persname')
+to_field 'creator_persname_ssim', extract_xpath('/ead/archdesc/did/origination/persname')
+to_field 'creator_corpname_ssm', extract_xpath('/ead/archdesc/did/origination/corpname')
+to_field 'creator_corpname_sim', extract_xpath('/ead/archdesc/did/origination/corpname')
+to_field 'creator_corpname_ssim', extract_xpath('/ead/archdesc/did/origination/corpname')
+to_field 'creator_famname_ssm', extract_xpath('/ead/archdesc/did/origination/famname')
+to_field 'creator_famname_ssim', extract_xpath('/ead/archdesc/did/origination/famname')
+
+to_field 'persname_sim', extract_xpath('//persname')
+
+to_field 'creators_ssim' do |_record, accumulator, context|
+  accumulator.concat context.output_hash['creator_persname_ssm'] if context.output_hash['creator_persname_ssm']
+  accumulator.concat context.output_hash['creator_corpname_ssm'] if context.output_hash['creator_corpname_ssm']
+  accumulator.concat context.output_hash['creator_famname_ssm'] if context.output_hash['creator_famname_ssm']
+end
+
+to_field 'places_sim', extract_xpath('/ead/archdesc/controlaccess/geogname')
+to_field 'places_ssim', extract_xpath('/ead/archdesc/controlaccess/geogname')
+to_field 'places_ssm', extract_xpath('/ead/archdesc/controlaccess/geogname')
+
+to_field 'access_terms_ssm', extract_xpath('/ead/archdesc/userestrict/*[local-name()!="head"]')
+
+to_field 'acqinfo_ssim', extract_xpath('/ead/archdesc/acqinfo/*[local-name()!="head"]')
+to_field 'acqinfo_ssim', extract_xpath('/ead/archdesc/descgrp/acqinfo/*[local-name()!="head"]')
+
+to_field 'access_subjects_ssim', extract_xpath('/ead/archdesc/controlaccess', to_text: false) do |_record, accumulator|
+  accumulator.map! do |element|
+    %w[subject function occupation genreform].map do |selector|
+      element.xpath(".//#{selector}").map(&:text)
+    end
+  end.flatten!
+end
+
+to_field 'access_subjects_ssm' do |_record, accumulator, context|
+  accumulator.concat Array.wrap(context.output_hash['access_subjects_ssim'])
+end
+
+to_field 'has_online_content_ssim', extract_xpath('.//dao') do |_record, accumulator|
+  accumulator.replace([accumulator.any?])
+end
+
+to_field 'digital_objects_ssm', extract_xpath('/ead/archdesc/did/dao|/ead/archdesc/dao', to_text: false) do |_record, accumulator|
+  accumulator.map! do |dao|
+    label = dao.attributes['title']&.value ||
+            dao.xpath('daodesc/p')&.text
+    href = (dao.attributes['href'] || dao.attributes['xlink:href'])&.value
+    Arclight::DigitalObject.new(label: label, href: href).to_json
+  end
+end
+
+to_field 'extent_ssm', extract_xpath('/ead/archdesc/did/physdesc/extent')
+to_field 'extent_teim', extract_xpath('/ead/archdesc/did/physdesc/extent')
+to_field 'genreform_sim', extract_xpath('/ead/archdesc/controlaccess/genreform')
+to_field 'genreform_ssm', extract_xpath('/ead/archdesc/controlaccess/genreform')
+
+to_field 'date_range_sim', extract_xpath('/ead/archdesc/did/unitdate/@normal', to_text: false) do |_record, accumulator|
+  range = Arclight::YearRange.new
+  next range.years if accumulator.blank?
+
+  ranges = accumulator.map(&:to_s)
+  range << range.parse_ranges(ranges)
+  accumulator.replace range.years
+end
+
+SEARCHABLE_NOTES_FIELDS.map do |selector|
+  to_field "#{selector}_ssm", extract_xpath("/ead/archdesc/#{selector}/*[local-name()!='head']", to_text: false)
+  to_field "#{selector}_heading_ssm", extract_xpath("/ead/archdesc/#{selector}/head") unless selector == 'prefercite'
+  to_field "#{selector}_teim", extract_xpath("/ead/archdesc/#{selector}/*[local-name()!='head']")
+end
+
+DID_SEARCHABLE_NOTES_FIELDS.map do |selector|
+  to_field "#{selector}_ssm", extract_xpath("/ead/archdesc/did/#{selector}", to_text: false)
+end
+
+NAME_ELEMENTS.map do |selector|
+  to_field 'names_coll_ssim', extract_xpath("/ead/archdesc/controlaccess/#{selector}")
+  to_field 'names_ssim', extract_xpath("//#{selector}")
+  to_field "#{selector}_ssm", extract_xpath("//#{selector}")
+end
+
+to_field 'corpname_sim', extract_xpath('//corpname')
+
+to_field 'language_sim', extract_xpath('/ead/archdesc/did/langmaterial')
+to_field 'language_ssm', extract_xpath('/ead/archdesc/did/langmaterial')
+
+to_field 'descrules_ssm', extract_xpath('/ead/eadheader/profiledesc/descrules')
+
+# =============================
+# Each component child document
+# <c> <c01> <c12>
+# =============================
+
+compose 'components', ->(record, accumulator, _context) { accumulator.concat record.xpath('//*[is_component(.)]', NokogiriXpathExtensions.new) } do
+  to_field 'ref_ssi' do |record, accumulator, context|
+    accumulator << if record.attribute('id').blank?
+                     strategy = Arclight::MissingIdStrategy.selected
+                     hexdigest = strategy.new(record).to_hexdigest
+                     parent_id = context.clipboard[:parent].output_hash['id'].first
+                     logger.warn('MISSING ID WARNING') do
+                       [
+                         "A component in #{parent_id} did not have an ID so one was minted using the #{strategy} strategy.",
+                         "The ID of this document will be #{parent_id}#{hexdigest}."
+                       ].join(' ')
+                     end
+                     record['id'] = hexdigest
+                     hexdigest
+                   else
+                     record.attribute('id')&.value&.strip&.gsub('.', '-')
+                   end
+  end
+  to_field 'ref_ssm' do |_record, accumulator, context|
+    accumulator.concat context.output_hash['ref_ssi']
+  end
+
+  to_field 'id' do |_record, accumulator, context|
+    accumulator << [
+      context.clipboard[:parent].output_hash['id'],
+      context.output_hash['ref_ssi']
+    ].join('')
+  end
+
+  to_field 'ead_ssi' do |_record, accumulator, context|
+    accumulator << context.clipboard[:parent].output_hash['ead_ssi'].first
+  end
+
+  to_field 'title_filing_si', extract_xpath('./did/unittitle'), first_only
+  to_field 'title_ssm', extract_xpath('./did/unittitle')
+  to_field 'title_teim', extract_xpath('./did/unittitle')
+
+  to_field 'unitdate_bulk_ssim', extract_xpath('./did/unitdate[@type="bulk"]')
+  to_field 'unitdate_inclusive_ssm', extract_xpath('./did/unitdate[@type="inclusive"]')
+  to_field 'unitdate_other_ssim', extract_xpath('./did/unitdate[not(@type)]')
+
+  to_field 'normalized_title_ssm' do |_record, accumulator, context|
+    dates = Arclight::NormalizedDate.new(
+      context.output_hash['unitdate_inclusive_ssm'],
+      context.output_hash['unitdate_bulk_ssim'],
+      context.output_hash['unitdate_other_ssim']
+    ).to_s
+    title = context.output_hash['title_ssm']&.first
+    accumulator << Arclight::NormalizedTitle.new(title, dates).to_s
+  end
+
+  to_field 'normalized_date_ssm' do |_record, accumulator, context|
+    accumulator << Arclight::NormalizedDate.new(
+      context.output_hash['unitdate_inclusive_ssm'],
+      context.output_hash['unitdate_bulk_ssim'],
+      context.output_hash['unitdate_other_ssim']
+    ).to_s
+  end
+
+  to_field 'component_level_isim' do |record, accumulator|
+    accumulator << 1 + NokogiriXpathExtensions.new.is_component(record.ancestors).count
+  end
+
+  to_field 'parent_ssim' do |record, accumulator, context|
+    accumulator << context.clipboard[:parent].output_hash['id'].first
+    accumulator.concat NokogiriXpathExtensions.new.is_component(record.ancestors).reverse.map { |n| n.attribute('id')&.value }
+  end
+
+  to_field 'parent_ssi' do |_record, accumulator, context|
+    accumulator << context.output_hash['parent_ssim'].last
+  end
+
+  to_field 'parent_unittitles_ssm' do |_rec, accumulator, context|
+    # top level document
+    accumulator.concat context.clipboard[:parent].output_hash['normalized_title_ssm']
+    parent_ssim = context.output_hash['parent_ssim']
+    components = context.clipboard[:parent].output_hash['components']
+
+    # other components
+    if parent_ssim && components
+      ancestors = parent_ssim.drop(1).map { |x| [x] }
+      accumulator.concat components.select { |c| ancestors.include? c['ref_ssi'] }.flat_map { |c| c['normalized_title_ssm'] }
+    end
+  end
+
+  to_field 'parent_unittitles_teim' do |_record, accumulator, context|
+    accumulator.concat context.output_hash['parent_unittitles_ssm']
+  end
+
+  to_field 'parent_levels_ssm' do |_record, accumulator, context|
+    ## Top level document
+    accumulator.concat context.clipboard[:parent].output_hash['level_ssm']
+    ## Other components
+    context.output_hash['parent_ssim']&.drop(1)&.each do |id|
+      accumulator.concat Array
+        .wrap(context.clipboard[:parent].output_hash['components'])
+        .select { |c| c['ref_ssi'] == [id] }.map { |c| c['level_ssm'] }.flatten
+    end
+  end
+
+  to_field 'unitid_ssm', extract_xpath('./did/unitid')
+  to_field 'collection_unitid_ssm' do |_record, accumulator, context|
+    accumulator.concat Array.wrap(context.clipboard[:parent].output_hash['unitid_ssm'])
+  end
+  to_field 'repository_ssm' do |_record, accumulator, context|
+    accumulator << context.clipboard[:parent].clipboard[:repository]
+  end
+  to_field 'repository_sim' do |_record, accumulator, context|
+    accumulator << context.clipboard[:parent].clipboard[:repository]
+  end
+  to_field 'collection_ssm' do |_record, accumulator, context|
+    accumulator.concat context.clipboard[:parent].output_hash['normalized_title_ssm']
+  end
+  to_field 'collection_sim' do |_record, accumulator, context|
+    accumulator.concat context.clipboard[:parent].output_hash['normalized_title_ssm']
+  end
+  to_field 'collection_ssi' do |_record, accumulator, context|
+    accumulator.concat context.clipboard[:parent].output_hash['normalized_title_ssm']
+  end
+
+  to_field 'extent_ssm', extract_xpath('./did/physdesc/extent')
+  to_field 'extent_teim', extract_xpath('./did/physdesc/extent')
+
+  to_field 'creator_ssm', extract_xpath('./did/origination')
+  to_field 'creator_ssim', extract_xpath('./did/origination')
+  to_field 'creators_ssim', extract_xpath('./did/origination')
+  to_field 'creator_sort' do |record, accumulator|
+    accumulator << record.xpath('./did/origination').map(&:text).join(', ')
+  end
+  to_field 'collection_creator_ssm' do |_record, accumulator, context|
+    accumulator.concat Array.wrap(context.clipboard[:parent].output_hash['creator_ssm'])
+  end
+  to_field 'has_online_content_ssim', extract_xpath('.//dao') do |_record, accumulator|
+    accumulator.replace([accumulator.any?])
+  end
+  to_field 'child_component_count_isim' do |record, accumulator|
+    accumulator << NokogiriXpathExtensions.new.is_component(record.children).count
+  end
+
+  to_field 'ref_ssm' do |record, accumulator|
+    accumulator << record.attribute('id')
+  end
+
+  to_field 'level_ssm' do |record, accumulator|
+    level = record.attribute('level')&.value
+    other_level = record.attribute('otherlevel')&.value
+    accumulator << Arclight::LevelLabel.new(level, other_level).to_s
+  end
+
+  to_field 'level_sim' do |_record, accumulator, context|
+    next unless context.output_hash['level_ssm']
+
+    accumulator.concat context.output_hash['level_ssm']&.map(&:capitalize)
+  end
+
+  to_field 'sort_ii' do |_record, accumulator, context|
+    accumulator.replace([context.position])
+  end
+
+  # Get the <accessrestrict> from the closest ancestor that has one (includes top-level)
+  to_field 'parent_access_restrict_ssm' do |record, accumulator|
+    accumulator.concat Array
+      .wrap(record.xpath('(./ancestor::*/accessrestrict)[last()]/*[local-name()!="head"]')
+      .map(&:text))
+  end
+
+  # Get the <userestrict> from self OR the closest ancestor that has one (includes top-level)
+  to_field 'parent_access_terms_ssm' do |record, accumulator|
+    accumulator.concat Array
+      .wrap(record.xpath('(./ancestor-or-self::*/userestrict)[last()]/*[local-name()!="head"]')
+      .map(&:text))
+  end
+
+  to_field 'digital_objects_ssm', extract_xpath('./dao|./did/dao', to_text: false) do |_record, accumulator|
+    accumulator.map! do |dao|
+      label = dao.attributes['title']&.value ||
+              dao.xpath('daodesc/p')&.text
+      href = (dao.attributes['href'] || dao.attributes['xlink:href'])&.value
+      Arclight::DigitalObject.new(label: label, href: href).to_json
+    end
+  end
+
+  to_field 'date_range_sim', extract_xpath('./did/unitdate/@normal', to_text: false) do |_record, accumulator|
+    range = Arclight::YearRange.new
+    next range.years if accumulator.blank?
+
+    ranges = accumulator.map(&:to_s)
+    range << range.parse_ranges(ranges)
+    accumulator.replace range.years
+  end
+
+  NAME_ELEMENTS.map do |selector|
+    to_field 'names_ssim', extract_xpath("./controlaccess/#{selector}")
+    to_field "#{selector}_ssm", extract_xpath(".//#{selector}")
+  end
+
+  to_field 'geogname_sim', extract_xpath('./controlaccess/geogname')
+  to_field 'geogname_ssm', extract_xpath('./controlaccess/geogname')
+  to_field 'places_ssim', extract_xpath('./controlaccess/geogname')
+
+  to_field 'access_subjects_ssim', extract_xpath('./controlaccess', to_text: false) do |_record, accumulator|
+    accumulator.map! do |element|
+      %w[subject function occupation genreform].map do |selector|
+        element.xpath(".//#{selector}").map(&:text)
+      end
+    end.flatten!
+  end
+
+  to_field 'access_subjects_ssm' do |_record, accumulator, context|
+    accumulator.concat(context.output_hash.fetch('access_subjects_ssim', []))
+  end
+
+  to_field 'acqinfo_ssim', extract_xpath('/ead/archdesc/acqinfo/*[local-name()!="head"]')
+  to_field 'acqinfo_ssim', extract_xpath('/ead/archdesc/descgrp/acqinfo/*[local-name()!="head"]')
+  to_field 'acqinfo_ssim', extract_xpath('./acqinfo/*[local-name()!="head"]')
+  to_field 'acqinfo_ssim', extract_xpath('./descgrp/acqinfo/*[local-name()!="head"]')
+
+  to_field 'language_ssm', extract_xpath('./did/langmaterial')
+  to_field 'containers_ssim' do |record, accumulator|
+    record.xpath('./did/container').each do |node|
+      accumulator << [node.attribute('type'), node.text].join(' ').strip
+    end
+  end
+
+  SEARCHABLE_NOTES_FIELDS.map do |selector|
+    to_field "#{selector}_ssm", extract_xpath("./#{selector}/*[local-name()!='head']", to_text: false)
+    to_field "#{selector}_heading_ssm", extract_xpath("./#{selector}/head")
+    to_field "#{selector}_teim", extract_xpath("./#{selector}/*[local-name()!='head']")
+  end
+  DID_SEARCHABLE_NOTES_FIELDS.map do |selector|
+    to_field "#{selector}_ssm", extract_xpath("./did/#{selector}", to_text: false)
+  end
+  to_field 'did_note_ssm', extract_xpath('./did/note')
+end
+
+each_record do |_record, context|
+  context.output_hash['components'] &&= context.output_hash['components'].select { |c| c.keys.any? }
+end
+
+##
+# Used for evaluating xpath components to find
+class NokogiriXpathExtensions
+  # rubocop:disable Naming/PredicateName, Style/FormatString
+  def is_component(node_set)
+    node_set.find_all do |node|
+      component_elements = (1..12).map { |i| "c#{'%02d' % i}" }
+      component_elements.push 'c'
+      component_elements.include? node.name
+    end
+  end
+  # rubocop:enable Naming/PredicateName, Style/FormatString
+end