bulkrax 3.0.0.beta3 → 3.0.0.beta4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 144e5a05fdf0d22faa759b2f8529ec971af9001f10c8bbe74d705d954b85e048
-  data.tar.gz: 569cdb00a7f533e5b0e234079d5e2ae0996f7c5230675fec158ce62c06a82987
+  metadata.gz: 29702685c19fb65a1cbf963615774541488af30e48feca79bdbf296edee7d4d8
+  data.tar.gz: f8b902f88a523155bab3d819905ffa25da29cad05178c445541a1159b84c94a9
 SHA512:
-  metadata.gz: 961ef2fcaccad86d47563de89538446a3757b54cd1abbaabe81e2479b902088ea62a06181fa22819e37df52cd6a32ce4a6c52892cd57ef89983b09a8bea403ef
-  data.tar.gz: 432704c4491eef71eb3a3bc4a294d13b9fa4e08b616c031a71510579919cdde96fa06d697f096016858725f6abf6d4a77414b2533f7aaeaa863e9031725f7ce5
+  metadata.gz: 3e6f041b8180b95d6f126649a0fe7da5a02f2a5afae88fd8f79e6f8c04bd9d4e854e4f626eb032132599ae865f8b4c8bdfb26a7f17469bef8426bc702bd88665
+  data.tar.gz: c4fcdb5b393970c9a79569a34e6317616ba96f9a953f5a963b413266621a76cb6a6f8c79de0a1f3b0753cd350e1ea11114d252a7724552b70f06ff4819657eb3

data/app/factories/bulkrax/object_factory.rb

@@ -7,10 +7,10 @@ module Bulkrax
     include DynamicRecordLookup
 
     define_model_callbacks :save, :create
-    attr_reader :attributes, :object, :source_identifier_value, :klass, :replace_files, :update_files, :work_identifier, :related_parents_parsed_mapping
+    attr_reader :attributes, :object, :source_identifier_value, :klass, :replace_files, :update_files, :work_identifier, :related_parents_parsed_mapping, :importer_run_id
 
     # rubocop:disable Metrics/ParameterLists
-    def initialize(attributes:, source_identifier_value:, work_identifier:, related_parents_parsed_mapping: nil, replace_files: false, user: nil, klass: nil, update_files: false)
+    def initialize(attributes:, source_identifier_value:, work_identifier:, related_parents_parsed_mapping: nil, replace_files: false, user: nil, klass: nil, importer_run_id: nil, update_files: false)
       @attributes = ActiveSupport::HashWithIndifferentAccess.new(attributes)
       @replace_files = replace_files
       @update_files = update_files
@@ -19,6 +19,7 @@ module Bulkrax
       @related_parents_parsed_mapping = related_parents_parsed_mapping
       @source_identifier_value = source_identifier_value
       @klass = klass || Bulkrax.default_work_type.constantize
+      @importer_run_id = importer_run_id
     end
     # rubocop:enable Metrics/ParameterLists
 
@@ -137,6 +138,7 @@ module Bulkrax
     end
 
     def create_collection(attrs)
+      attrs = clean_attrs(attrs)
       attrs = collection_type(attrs)
       persist_collection_memberships(parent: find_collection(attributes[related_parents_parsed_mapping]), child: object) if attributes[related_parents_parsed_mapping].present?
       object.attributes = attrs
@@ -152,8 +154,9 @@ module Bulkrax
 
     # This method is heavily inspired by Hyrax's AttachFilesToWorkJob
     def create_file_set(attrs)
-      work = find_record(attributes[related_parents_parsed_mapping].first)
+      _, work = find_record(attributes[related_parents_parsed_mapping].first, importer_run_id)
       work_permissions = work.permissions.map(&:to_hash)
+      attrs = clean_attrs(attrs)
       file_set_attrs = attrs.slice(*object.attributes.keys)
       object.assign_attributes(file_set_attrs)
 
@@ -202,8 +205,16 @@ module Bulkrax
       end
     end
 
+    def clean_attrs(attrs)
+      # avoid the "ArgumentError: Identifier must be a string of size > 0 in order to be treeified" error
+      # when setting object.attributes
+      attrs.delete('id') if attrs['id'].blank?
+      attrs
+    end
+
     def collection_type(attrs)
       return attrs if attrs['collection_type_gid'].present?
+
       attrs['collection_type_gid'] = Hyrax::CollectionType.find_or_create_default_collection_type.gid
       attrs
     end

data/app/jobs/bulkrax/create_relationships_job.rb

@@ -37,11 +37,11 @@ module Bulkrax
       end.sort_by(&:order)
 
       @importer_run_id = importer_run_id
-      @parent_record = find_record(parent_identifier)
+      @parent_entry, @parent_record = find_record(parent_identifier, importer_run_id)
       @child_records = { works: [], collections: [] }
       pending_relationships.each do |rel|
         raise ::StandardError, %("#{rel}" needs either a child or a parent to create a relationship) if rel.child_id.nil? || rel.parent_id.nil?
-        child_record = find_record(rel.child_id)
+        _, child_record = find_record(rel.child_id, importer_run_id)
         child_record.is_a?(::Collection) ? @child_records[:collections] << child_record : @child_records[:works] << child_record
       end
 
@@ -53,9 +53,9 @@ module Bulkrax
         return false # stop current job from continuing to run after rescheduling
       end
 
-      @parent_entry = Bulkrax::Entry.where(identifier: parent_identifier,
-                                           importerexporter_id: ImporterRun.find(importer_run_id).importer_id,
-                                           importerexporter_type: "Bulkrax::Importer").first
+      @parent_entry ||= Bulkrax::Entry.where(identifier: parent_identifier,
+                                             importerexporter_id: ImporterRun.find(importer_run_id).importer_id,
+                                             importerexporter_type: "Bulkrax::Importer").first
       create_relationships
       pending_relationships.each(&:destroy)
     rescue ::StandardError => e
@@ -91,7 +91,8 @@ module Bulkrax
           related_parents_parsed_mapping: parent_entry.parser.related_parents_parsed_mapping,
           replace_files: false,
           user: user,
-          klass: child_record.class
+          klass: child_record.class,
+          importer_run_id: importer_run_id
         ).run
         # TODO: add counters for :processed_parents and :failed_parents
         Bulkrax::ImporterRun.find(importer_run_id).increment!(:processed_relationships) # rubocop:disable Rails/SkipsModelValidations
@@ -109,7 +110,8 @@ module Bulkrax
         related_parents_parsed_mapping: parent_entry.parser.related_parents_parsed_mapping,
         replace_files: false,
         user: user,
-        klass: parent_record.class
+        klass: parent_record.class,
+        importer_run_id: importer_run_id
       ).run
       # TODO: add counters for :processed_parents and :failed_parents
       Bulkrax::ImporterRun.find(importer_run_id).increment!(:processed_relationships) # rubocop:disable Rails/SkipsModelValidations
@@ -132,7 +134,8 @@ module Bulkrax
         related_parents_parsed_mapping: parent_entry.parser.related_parents_parsed_mapping,
         replace_files: false,
         user: user,
-        klass: parent_record.class
+        klass: parent_record.class,
+        importer_run_id: importer_run_id
       ).run
       # TODO: add counters for :processed_parents and :failed_parents
       Bulkrax::ImporterRun.find(importer_run_id).increment!(:processed_relationships) # rubocop:disable Rails/SkipsModelValidations

data/app/jobs/bulkrax/import_file_set_job.rb

@@ -7,7 +7,10 @@ module Bulkrax
 
     queue_as :import
 
+    attr_reader :importer_run_id
+
     def perform(entry_id, importer_run_id)
+      @importer_run_id = importer_run_id
       entry = Entry.find(entry_id)
       parent_identifier = entry.raw_metadata[entry.related_parents_raw_mapping]&.strip
 
@@ -63,7 +66,8 @@ module Bulkrax
     end
 
     def find_parent_record(parent_identifier)
-      @parent_record ||= find_record(parent_identifier)
+      @parent_record ||= find_record(parent_identifier, importer_run_id)
+      @parent_record = parent_record.last if parent_record.is_a? Array
     end
   end
 end

data/app/models/bulkrax/csv_entry.rb

@@ -41,6 +41,8 @@ module Bulkrax
       self.parsed_metadata = {}
       add_identifier
       add_ingested_metadata
+      # TODO(alishaevn): remove the collections stuff entirely and only reference collections via the new parents code
+      add_collections
       add_visibility
       add_metadata_for_model
       add_rights_statement
@@ -241,16 +243,17 @@ module Bulkrax
     def possible_collection_ids
       return @possible_collection_ids if @possible_collection_ids.present?
 
-      collection_field_mapping = self.class.parent_field(parser)
-      return [] unless collection_field_mapping.present? && record[collection_field_mapping].present?
+      parent_field_mapping = self.class.parent_field(parser)
+      return [] unless parent_field_mapping.present? && record[parent_field_mapping].present?
 
       identifiers = []
-      split_titles = record[collection_field_mapping].split(/\s*[;|]\s*/)
-      split_titles.each do |c_title|
-        matching_collection_entries = importerexporter.entries.select { |e| e.raw_metadata['title'] == c_title }
+      split_references = record[parent_field_mapping].split(/\s*[;|]\s*/)
+      split_references.each do |c_reference|
+        matching_collection_entries = importerexporter.entries.select { |e| e.raw_metadata[work_identifier] == c_reference && e.is_a?(CsvCollectionEntry) }
         raise ::StandardError, 'Only expected to find one matching entry' if matching_collection_entries.count > 1
         identifiers << matching_collection_entries.first&.identifier
       end
+
       @possible_collection_ids = identifiers.compact.presence || []
     end
 
@@ -267,6 +270,7 @@ module Bulkrax
           self.collection_ids << c.id unless skip
         end
       end
+
       self.collection_ids
     end
 

data/app/models/concerns/bulkrax/dynamic_record_lookup.rb

@@ -6,9 +6,15 @@ module Bulkrax
     # has the provided identifier.
     #
     # @param identifier [String] Work/Collection ID or Bulkrax::Entry source_identifier
-    # @return [Work, Collection, nil] Work or Collection if found, otherwise nil
-    def find_record(identifier)
-      record = Entry.find_by(identifier: identifier)
+    # @param importer_run_id [Number] ID of the current_run of this Importer Job
+    # @return [Entry, nil], [Work, Collection, nil] Entry if found, otherwise nil and a Work or Collection if found, otherwise nil
+    def find_record(identifier, importer_run_id = nil)
+      # check for our entry in our current importer first
+      importer_id = ImporterRun.find(importer_run_id).importer_id
+      record = Entry.find_by(identifier: identifier, importerexporter_id: importer_id) || Entry.find_by(identifier: identifier)
+
+      # TODO(alishaevn): discuss whether we are only looking for Collection models here
+      # use ActiveFedora::Base.find(identifier) instead?
       record ||= ::Collection.where(id: identifier).first # rubocop:disable Rails/FindBy
       if record.blank?
         available_work_types.each do |work_type|
@@ -16,7 +22,9 @@ module Bulkrax
         end
       end
 
-      record.is_a?(Entry) ? record.factory.find : record
+      # return the found entry here instead of searching for it again in the CreateRelationshipsJob
+      # also accounts for when the found entry isn't a part of this importer
+      record.is_a?(Entry) ? [record, record.factory.find] : [nil, record]
     end
 
     # Check if the record is a Work

data/app/models/concerns/bulkrax/file_factory.rb

@@ -45,6 +45,8 @@ module Bulkrax
     end
 
     def new_remote_files
+      return if object.is_a? FileSet
+
       @new_remote_files ||= if object.present? && object.file_sets.present?
                               parsed_remote_files.select do |file|
                                 # is the url valid?
@@ -101,7 +103,12 @@ module Bulkrax
     end
 
     def local_file_sets
-      @local_file_sets ||= object&.ordered_file_sets
+      @local_file_sets ||= ordered_file_sets
+    end
+
+    def ordered_file_sets
+      # OVERRIDE Hyrda-works 1.2.0 - this method was deprecated in v1.0
+      object&.ordered_members.to_a.select(&:file_set?)
     end
 
     def import_files

data/app/models/concerns/bulkrax/import_behavior.rb

@@ -98,6 +98,7 @@ module Bulkrax
                                               replace_files: replace_files,
                                               user: user,
                                               klass: factory_class,
+                                              importer_run_id: importerexporter.last_run.id,
                                               update_files: update_files)
     end
 

data/app/views/bulkrax/importers/show.html.erb

@@ -97,13 +97,8 @@
               <% @work_entries.each do |e| %>
                 <tr>
                   <td><%= link_to e.identifier, bulkrax.importer_entry_path(@importer.id, e.id) %></td>
-                  <% if e.parsed_metadata.present? && e.parsed_metadata.dig("collections").present? %>
-                    <td><%= e.parsed_metadata.dig("collections").map {|c| c['id'] }.join('; ') %></td>
-                  <% elsif e.raw_metadata.present? %>
-                    <td><%= Array.wrap(e.raw_metadata.dig("collection")).join(';') %></td>
-                  <% else %>
-                    <td></td>
-                  <% end %>
+                  <% collection_titles = e.collection_ids.map {|id| c = Collection.find id; c.title.first } %>
+                  <td><%= collection_titles.join('; ') %></td>
                   <td><%= e.id %></td>
                   <% if e.status == "Complete" %>
                     <td><span class="glyphicon glyphicon-ok" style="color: green;"></span> <%= e.status %></td>

data/lib/bulkrax/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Bulkrax
-  VERSION = '3.0.0.beta3'
+  VERSION = '3.0.0.beta4'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: bulkrax
 version: !ruby/object:Gem::Version
-  version: 3.0.0.beta3
+  version: 3.0.0.beta4
 platform: ruby
 authors:
 - Rob Kaufman
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2022-03-30 00:00:00.000000000 Z
+date: 2022-04-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rails