bulkrax 2.2.4 → 2.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 9ca4838e8fe86d901fd9196ab8c1a84f7072fae590a1fc411f8ee057756f9db9
-  data.tar.gz: 8b6dde2983bac873a5dbbd0f70aab512199344f00a2a7cbdcbea5f5b2a9fe4c5
+  metadata.gz: 264fedacd4fc13cbd7450068ff6f16317b9ff33e6f28edf48484652fcdf50c0a
+  data.tar.gz: c91b6d74984569513ae567cc879558fbbc5374b6379ca272c744d5be5972ef45
 SHA512:
-  metadata.gz: 14ee5cdf3bfee220faebd377b6ce25979397c4b72ac9de3f10d92d13f57d60fffecb089e26f7bdc1c5ccb147430fb637635053e0a601a1f0f96ad2082d1d10b9
-  data.tar.gz: 0236d3b24309e7a0338d1fb38d40e0ab852ff452514311c5e7711fc8efcc39a19a48a044622b2168d43d1fd1517cd6a61a87273af350ab0dd91c6576230911d6
+  metadata.gz: 501a1a4bae256c70c35524e3a41d46f25c4d464c180355b10ae47ea13acc1ee20c22a1763c4908ce3821162b7ebd7c1c3a4b975f27ce32a9014a50d04c1c1f18
+  data.tar.gz: 63fbe35c7bf2496d46434fc1d1b8775e4d1b3c5c43529ed258d030ec6816d8a89ad7c9e90a9dc73e21e699f7efabee1dc6279d59840defce087eeeb7db023483

data/app/models/bulkrax/csv_entry.rb

@@ -3,7 +3,10 @@
 require 'csv'
 
 module Bulkrax
-  class CsvEntry < Entry
+  # TODO: We need to rework this class some to address the Metrics/ClassLength rubocop offense.
+  # We do too much in these entry classes. We need to extract the common logic from the various
+  # entry models into a module that can be shared between them.
+  class CsvEntry < Entry # rubocop:disable Metrics/ClassLength
     serialize :raw_metadata, JSON
 
     def self.fields_from_data(data)
@@ -229,10 +232,19 @@ module Bulkrax
         'Creating Collections using the collection_field_mapping will no longer be supported as of Bulkrax version 3.0.' \
         ' Please configure Bulkrax to use related_parents_field_mapping and related_children_field_mapping instead.'
       )
-      @possible_collection_ids ||= record.inject([]) do |memo, (key, value)|
-        memo += value.split(/\s*[:;|]\s*/) if self.class.collection_field.to_s == key_without_numbers(key) && value.present?
-        memo
-      end || []
+      return @possible_collection_ids if @possible_collection_ids.present?
+
+      collection_field_mapping = self.class.collection_field
+      return [] unless collection_field_mapping.present? && record[collection_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 }
+        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
 
     def collections_created?

data/app/parsers/bulkrax/csv_parser.rb

@@ -88,31 +88,30 @@ module Bulkrax
       collections.each_with_index do |collection, index|
         next if collection.blank?
         break if records.find_index(collection).present? && limit_reached?(limit, records.find_index(collection))
-        ActiveSupport::Deprecation.warn(
-          'Creating Collections using the collection_field_mapping will no longer be supported as of Bulkrax version 3.0.' \
-          ' Please configure Bulkrax to use related_parents_field_mapping and related_children_field_mapping instead.'
-        )
 
         ## BEGIN
         # Add required metadata to collections being imported using the collection_field_mapping, which only have a :title
         # TODO: Remove once collection_field_mapping is removed
-        metadata = if collection.delete(:from_collection_field_mapping)
-                     uci = unique_collection_identifier(collection)
-                     {
-                       title: collection[:title],
-                       work_identifier => uci,
-                       source_identifier => uci,
-                       visibility: 'open',
-                       collection_type_gid: ::Hyrax::CollectionType.find_or_create_default_collection_type.gid
-                     }
-                   end
+        metadata = add_required_collection_metadata(collection)
         collection_hash = metadata.presence || collection
         ## END
 
         new_entry = find_or_create_entry(collection_entry_class, collection_hash[source_identifier], 'Bulkrax::Importer', collection_hash)
         increment_counters(index, collection: true)
         # TODO: add support for :delete option
-        ImportCollectionJob.perform_now(new_entry.id, current_run.id)
+        if collection.key?(:from_collection_field_mapping)
+          ActiveSupport::Deprecation.warn(
+            'Creating Collections using the collection_field_mapping will no longer be supported as of Bulkrax version 3.0.' \
+            ' Please configure Bulkrax to use related_parents_field_mapping and related_children_field_mapping instead.'
+          )
+          # When importing collections using the deprecated collection_field_mapping, the collection MUST be created
+          # before the work, so we use #perform_now to make sure that happens. The downside is, if a collection fails
+          # to import, it will stop the rest of the collections from importing successfully.
+          # TODO: Remove once collection_field_mapping is removed
+          ImportCollectionJob.perform_now(new_entry.id, current_run.id)
+        else
+          ImportCollectionJob.perform_later(new_entry.id, current_run.id)
+        end
       end
       importer.record_status
     rescue StandardError => e
@@ -152,6 +151,25 @@ module Bulkrax
       status_info(e)
     end
 
+    # Add required metadata to collections being imported using the collection_field_mapping, which only have a :title
+    # TODO: Remove once collection_field_mapping is removed
+    def add_required_collection_metadata(raw_collection_data)
+      return unless raw_collection_data.key?(:from_collection_field_mapping)
+      ActiveSupport::Deprecation.warn(
+        'Creating Collections using the collection_field_mapping will no longer be supported as of Bulkrax version 3.0.' \
+        ' Please configure Bulkrax to use related_parents_field_mapping and related_children_field_mapping instead.'
+      )
+
+      uci = unique_collection_identifier(raw_collection_data)
+      {
+        title: raw_collection_data[:title],
+        work_identifier => uci,
+        source_identifier => uci,
+        visibility: 'open',
+        collection_type_gid: ::Hyrax::CollectionType.find_or_create_default_collection_type.gid
+      }
+    end
+
     def write_partial_import_file(file)
       import_filename = import_file_path.split('/').last
       partial_import_filename = "#{File.basename(import_filename, '.csv')}_corrected_entries.csv"
@@ -188,13 +206,13 @@ module Bulkrax
 
       case importerexporter.export_from
       when 'all'
-        @work_ids = ActiveFedora::SolrService.query("has_model_ssim:(#{Hyrax.config.curation_concerns.join(' OR ')}) #{extra_filters}", rows: 2_147_483_647).map(&:id)
-        @collection_ids = ActiveFedora::SolrService.query("has_model_ssim:Collection #{extra_filters}", rows: 2_147_483_647).map(&:id)
-        @file_set_ids = ActiveFedora::SolrService.query("has_model_ssim:FileSet #{extra_filters}", rows: 2_147_483_647).map(&:id)
+        @work_ids = ActiveFedora::SolrService.query("has_model_ssim:(#{Hyrax.config.curation_concerns.join(' OR ')}) #{extra_filters}", method: :post, rows: 2_147_483_647).map(&:id)
+        @collection_ids = ActiveFedora::SolrService.query("has_model_ssim:Collection #{extra_filters}", method: :post, rows: 2_147_483_647).map(&:id)
+        @file_set_ids = ActiveFedora::SolrService.query("has_model_ssim:FileSet #{extra_filters}", method: :post, rows: 2_147_483_647).map(&:id)
       when 'collection'
-        @work_ids = ActiveFedora::SolrService.query("member_of_collection_ids_ssim:#{importerexporter.export_source + extra_filters}", rows: 2_000_000_000).map(&:id)
+        @work_ids = ActiveFedora::SolrService.query("member_of_collection_ids_ssim:#{importerexporter.export_source + extra_filters}", method: :post, rows: 2_000_000_000).map(&:id)
       when 'worktype'
-        @work_ids = ActiveFedora::SolrService.query("has_model_ssim:#{importerexporter.export_source + extra_filters}", rows: 2_000_000_000).map(&:id)
+        @work_ids = ActiveFedora::SolrService.query("has_model_ssim:#{importerexporter.export_source + extra_filters}", method: :post, rows: 2_000_000_000).map(&:id)
       when 'importer'
         set_ids_for_exporting_from_importer
       end
@@ -214,7 +232,7 @@ module Bulkrax
       extra_filters = extra_filters.presence || '*:*'
 
       { :@work_ids => ::Hyrax.config.curation_concerns, :@collection_ids => [::Collection], :@file_set_ids => [::FileSet] }.each do |instance_var, models_to_search|
-        instance_variable_set(instance_var, ActiveFedora::SolrService.get(
+        instance_variable_set(instance_var, ActiveFedora::SolrService.post(
           extra_filters.to_s,
           fq: [
             "#{work_identifier}_sim:(#{complete_entry_identifiers.join(' OR ')})",

data/lib/bulkrax/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Bulkrax
-  VERSION = '2.2.4'
+  VERSION = '2.3.0'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: bulkrax
 version: !ruby/object:Gem::Version
-  version: 2.2.4
+  version: 2.3.0
 platform: ruby
 authors:
 - Rob Kaufman
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2022-02-25 00:00:00.000000000 Z
+date: 2022-03-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rails