bulkrax 3.4.0 → 4.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f0ee151bc10b7485eb716463b2c4895165d6df9d73c3dd60813d2eb4de8161d1
-  data.tar.gz: f4e5ddfb5ac602eb20a85850f8ec5a9286a9a668f8c83548dd049db4d91a2a0e
+  metadata.gz: eb56d86ee90ae9e1cf0628504694e1301ab8f2d6b24ffa8fd323f8953a8ee956
+  data.tar.gz: 71056b077e300f27eee3bcccd9d7e2bee2fc7bdf2fc6ba9248b69a29f3994f9c
 SHA512:
-  metadata.gz: a5b029da7feaee11c8a3eb58e0c7150abcb06b6ca08f9d102134d5e1c9eef3049ae85b0084d74934b7504688da7d4958a0dc425c705b42101c1a3ce62a57d0c7
-  data.tar.gz: c807b68265c0d88b9e7faea4f6efdf94b3bcf90e965f5cc97cc259202bc55db976944eea8e4bf99876723f881d29d2fa7004dbd9985f7f8164648df517722133
+  metadata.gz: 05ea49e6f2c5e73cbddacf35dcaf9de499760d7093e3ae8f3ce4ea5ab28e25d065b7877607436fcbe02d21e17c2df940b0224f1ea7d638a602486ce807d99981
+  data.tar.gz: ecdda29924e09793e62684f16ebcd79cd90ab9e8204d011b89a577ca667a644709ff2df5b525ac6c32355426038c05d9fc3d74c6efb20ee7cfab653d9b89b67a

data/README.md

@@ -70,7 +70,7 @@ Bulkrax.setup do |config|
 end
 ```
 
-The [configuration guide](https://github.com/samvera-labs/bulkrax/wiki/Configuration) provides detailed instructions on the various available configurations.
+The [configuration guide](https://github.com/samvera-labs/bulkrax/wiki/Configuring-Bulkrax) provides detailed instructions on the various available configurations.
 
 Example:
 
@@ -120,7 +120,7 @@ It's unlikely that the incoming import data has fields that exactly match those
 
 By default, a mapping for the OAI parser has been added to map standard oai_dc fields to Hyrax basic_metadata. The other parsers have no default mapping, and will map any incoming fields to Hyrax properties with the same name. Configurations can be added in `config/intializers/bulkrax.rb`
 
-Configuring field mappings is documented in the [Bulkrax Configuration Guide](https://github.com/samvera-labs/bulkrax/wiki/Configuration).
+Configuring field mappings is documented in the [Bulkrax Configuration Guide](https://github.com/samvera-labs/bulkrax/wiki/Configuring-Bulkrax).
 
 ## Importing Files
 
@@ -151,7 +151,7 @@ end
 
 ## Customizing Bulkrax
 
-For further information on how to extend and customize Bulkrax, please see the [Bulkrax Customization Guide](https://github.com/samvera-labs/bulkrax/wiki/Customizing).
+For further information on how to extend and customize Bulkrax, please see the [Bulkrax Customization Guide](https://github.com/samvera-labs/bulkrax/wiki/Customizing-Bulkrax).
 
 ## How it Works
 Once you have Bulkrax installed, you will have access to an easy to use interface with which you are able to create, edit, delete, run, and re-run imports and exports.
@@ -191,8 +191,6 @@ We encourage everyone to help improve this project.  Bug reports and pull reques
 
 This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the [Contributor Covenant](https://contributor-covenant.org) code of conduct.
 
-All Contributors should have signed the Samvera Contributor License Agreement (CLA)
-
 ## Questions
 Questions can be sent to support@notch8.com. Please make sure to include "Bulkrax" in the subject line of your email.
 

data/app/controllers/bulkrax/exporters_controller.rb

@@ -127,7 +127,7 @@ module Bulkrax
     # Download methods
 
     def file_path
-      @exporter.exporter_export_zip_path
+      "#{@exporter.exporter_export_zip_path}/#{params['exporter']['exporter_export_zip_files']}"
     end
   end
 end

data/app/jobs/bulkrax/create_relationships_job.rb

@@ -42,10 +42,12 @@ module Bulkrax
       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_entry, 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
+        if child_record
+          child_record.is_a?(::Collection) ? @child_records[:collections] << child_record : @child_records[:works] << child_record
+        end
       end
 
-      if (child_records[:collections].blank? && child_records[:works].blank?) || parent_record.blank?
+      if (child_records[:collections].blank? && child_records[:works].blank?) || parent_record.nil?
         reschedule({ parent_identifier: parent_identifier, importer_run_id: importer_run_id })
         return false # stop current job from continuing to run after rescheduling
       end

data/app/models/bulkrax/entry.rb

@@ -4,8 +4,6 @@ module Bulkrax
   # Custom error class for collections_created?
   class CollectionsCreatedError < RuntimeError; end
   class OAIError < RuntimeError; end
-  # TODO: remove when ApplicationParser#bagit_zip_file_size_check is removed
-  class BagitZipError < RuntimeError; end
   class Entry < ApplicationRecord
     include Bulkrax::HasMatchers
     include Bulkrax::ImportBehavior

data/app/models/bulkrax/exporter.rb

@@ -124,9 +124,13 @@ module Bulkrax
     end
 
     def exporter_export_zip_path
-      @exporter_export_zip_path ||= File.join(parser.base_path('export'), "export_#{self.id}_#{self.exporter_runs.last.id}.zip")
+      @exporter_export_zip_path ||= File.join(parser.base_path('export'), "export_#{self.id}_#{self.exporter_runs.last.id}")
     rescue
-      @exporter_export_zip_path ||= File.join(parser.base_path('export'), "export_#{self.id}_0.zip")
+      @exporter_export_zip_path ||= File.join(parser.base_path('export'), "export_#{self.id}_0")
+    end
+
+    def exporter_export_zip_files
+      @exporter_export_zip_files ||= Dir["#{exporter_export_zip_path}/**"].map { |zip| Array(zip.split('/').last) }
     end
 
     def export_properties
@@ -137,5 +141,14 @@ module Bulkrax
     def metadata_only?
       export_type == 'metadata'
     end
+
+    def sort_zip_files(zip_files)
+      zip_files.sort_by do |item|
+        number = item.split('_').last.match(/\d+/)&.[](0) || 0.to_s
+        sort_number = number.rjust(4, "0")
+
+        sort_number
+      end
+    end
   end
 end

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

@@ -12,15 +12,14 @@ module Bulkrax
       # check for our entry in our current importer first
       importer_id = ImporterRun.find(importer_run_id).importer_id
       default_scope = { identifier: identifier, importerexporter_type: 'Bulkrax::Importer' }
-      record = Entry.find_by(default_scope.merge({ importerexporter_id: importer_id })) || Entry.find_by(default_scope)
 
-      # 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|
-          record ||= work_type.where(id: identifier).first # rubocop:disable Rails/FindBy
-        end
+      begin
+        # the identifier parameter can be a :source_identifier or the id of an object
+        record = Entry.find_by(default_scope.merge({ importerexporter_id: importer_id })) || Entry.find_by(default_scope)
+        record ||= ActiveFedora::Base.find(identifier)
+      # NameError for if ActiveFedora isn't installed
+      rescue NameError, ActiveFedora::ObjectNotFoundError
+        record = nil
       end
 
       # return the found entry here instead of searching for it again in the CreateRelationshipsJob

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

@@ -7,9 +7,6 @@ module Bulkrax
 
     def build_for_exporter
       build_export_metadata
-      # TODO(alishaevn): determine if the line below is still necessary
-      # the csv and bagit parsers also have write_files methods
-      write_files if export_type == 'full' && !importerexporter.parser_klass.include?('Bagit')
     rescue RSolr::Error::Http, CollectionsCreatedError => e
       raise e
     rescue StandardError => e
@@ -26,25 +23,6 @@ module Bulkrax
       @hyrax_record ||= ActiveFedora::Base.find(self.identifier)
     end
 
-    def write_files
-      return if hyrax_record.is_a?(Collection)
-
-      file_sets = hyrax_record.file_set? ? Array.wrap(hyrax_record) : hyrax_record.file_sets
-      file_sets << hyrax_record.thumbnail if hyrax_record.thumbnail.present? && hyrax_record.work? && exporter.include_thumbnails
-      file_sets.each do |fs|
-        path = File.join(exporter_export_path, 'files')
-        FileUtils.mkdir_p(path)
-        file = filename(fs)
-        require 'open-uri'
-        io = open(fs.original_file.uri)
-        next if file.blank?
-        File.open(File.join(path, file), 'wb') do |f|
-          f.write(io.read)
-          f.close
-        end
-      end
-    end
-
     # Prepend the file_set id to ensure a unique filename and also one that is not longer than 255 characters
     def filename(file_set)
       return if file_set.original_file.blank?

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

@@ -8,10 +8,14 @@ module Bulkrax
 
     def add_path_to_file
       parsed_metadata['file'].each_with_index do |filename, i|
-        path_to_file = ::File.join(parser.path_to_files, filename)
+        next if filename.blank?
+
+        path_to_file = parser.path_to_files(filename: filename)
 
         parsed_metadata['file'][i] = path_to_file
       end
+      parsed_metadata['file'].delete('')
+
       raise ::StandardError, "one or more file paths are invalid: #{parsed_metadata['file'].join(', ')}" unless parsed_metadata['file'].map { |file_path| ::File.file?(file_path) }.all?
 
       parsed_metadata['file']

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

@@ -12,8 +12,8 @@ module Bulkrax
           raise CollectionsCreatedError unless collections_created?
           @item = factory.run!
           add_user_to_permission_templates! if self.class.to_s.include?("Collection")
-          parent_jobs if self.parsed_metadata[related_parents_parsed_mapping].present?
-          child_jobs if self.parsed_metadata[related_children_parsed_mapping].present?
+          parent_jobs if self.parsed_metadata[related_parents_parsed_mapping]&.join.present?
+          child_jobs if self.parsed_metadata[related_children_parsed_mapping]&.join.present?
         end
       rescue RSolr::Error::Http, CollectionsCreatedError => e
         raise e

data/app/parsers/bulkrax/application_parser.rb

@@ -247,8 +247,6 @@ module Bulkrax
     def write
       write_files
       zip
-      # uncomment next line to debug for faulty zipping during bagit export
-      bagit_zip_file_size_check if importerexporter.parser_klass.include?('Bagit')
     end
 
     def unzip(file_to_unzip)
@@ -262,30 +260,13 @@ module Bulkrax
     end
 
     def zip
-      FileUtils.rm_rf(exporter_export_zip_path)
-      Zip::File.open(exporter_export_zip_path, create: true) do |zip_file|
-        Dir["#{exporter_export_path}/**/**"].each do |file|
-          zip_file.add(file.sub("#{exporter_export_path}/", ''), file)
-        end
-      end
-    end
+      FileUtils.mkdir_p(exporter_export_zip_path)
 
-    # TODO: remove Entry::BagitZipError as well as this method when we're sure it's not needed
-    def bagit_zip_file_size_check
-      Zip::File.open(exporter_export_zip_path) do |zip_file|
-        zip_file.select { |entry| entry.name.include?('data/') && entry.file? }.each do |zipped_file|
-          Dir["#{exporter_export_path}/**/data/*"].select { |file| file.include?(zipped_file.name) }.each do |file|
-            begin
-              raise BagitZipError, "Invalid Bag, file size mismatch for #{file.sub("#{exporter_export_path}/", '')}" if File.size(file) != zipped_file.size
-            rescue BagitZipError => e
-              matched_entry_ids = importerexporter.entry_ids.select do |id|
-                Bulkrax::Entry.find(id).identifier.include?(zipped_file.name.split('/').first)
-              end
-              matched_entry_ids.each do |entry_id|
-                Bulkrax::Entry.find(entry_id).status_info(e)
-                status_info('Complete (with failures)')
-              end
-            end
+      Dir["#{exporter_export_path}/**"].each do |folder|
+        zip_path = "#{exporter_export_zip_path.split('/').last}_#{folder.split('/').last}.zip"
+        Zip::File.open(File.join("#{exporter_export_zip_path}/#{zip_path}"), create: true) do |zip_file|
+          Dir["#{folder}/**/**"].each do |file|
+            zip_file.add(file.sub("#{folder}/", ''), file)
           end
         end
       end

data/app/parsers/bulkrax/bagit_parser.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 
 module Bulkrax
-  class BagitParser < ApplicationParser # rubocop:disable Metrics/ClassLength
+  class BagitParser < CsvParser # rubocop:disable Metrics/ClassLength
     include ExportBehavior
 
     def self.export_supported?
@@ -20,12 +20,8 @@ module Bulkrax
       rdf_format ? RdfEntry : CsvEntry
     end
 
-    def collection_entry_class
-      CsvCollectionEntry
-    end
-
-    def file_set_entry_class
-      CsvFileSetEntry
+    def path_to_files(filename:)
+      @path_to_files ||= Dir.glob(File.join(import_file_path, '**/data', filename)).first
     end
 
     # Take a random sample of 10 metadata_paths and work out the import fields from that
@@ -36,39 +32,41 @@ module Bulkrax
       end.flatten.compact.uniq
     end
 
-    # Assume a single metadata record per path
-    # Create an Array of all metadata records, one per file
+    # Create an Array of all metadata records
     def records(_opts = {})
       raise StandardError, 'No BagIt records were found' if bags.blank?
       @records ||= bags.map do |bag|
         path = metadata_path(bag)
         raise StandardError, 'No metadata files were found' if path.blank?
         data = entry_class.read_data(path)
-        data = entry_class.data_for_entry(data, source_identifier, self)
-        data[:file] = bag.bag_files.join('|') unless importerexporter.metadata_only?
-        data
+        get_data(bag, data)
       end
+
+      @records = @records.flatten
     end
 
-    # Find or create collections referenced by works
-    # If the import data also contains records for these works, they will be updated
-    # during create works
-    def create_collections
-      collections.each_with_index do |collection, index|
-        next if collection.blank?
-        metadata = {
-          title: [collection],
-          work_identifier => [collection],
-          visibility: 'open',
-          collection_type_gid: Hyrax::CollectionType.find_or_create_default_collection_type.gid
-        }
-        new_entry = find_or_create_entry(collection_entry_class, collection, 'Bulkrax::Importer', metadata)
-        ImportCollectionJob.perform_now(new_entry.id, current_run.id)
-        increment_counters(index, collection: true)
+    def get_data(bag, data)
+      if entry_class == CsvEntry
+        data = data.map do |data_row|
+          record_data = entry_class.data_for_entry(data_row, source_identifier, self)
+          next record_data if importerexporter.metadata_only?
+
+          record_data[:file] = bag.bag_files.join('|') if ::Hyrax.config.curation_concerns.include? record_data[:model]&.constantize
+          record_data
+        end
+      else
+        data = entry_class.data_for_entry(data, source_identifier, self)
+        data[:file] = bag.bag_files.join('|') unless importerexporter.metadata_only?
       end
+
+      data
     end
 
     def create_works
+      entry_class == CsvEntry ? super : create_rdf_works
+    end
+
+    def create_rdf_works
       records.each_with_index do |record, index|
         next unless record_has_source_identifier(record, index)
         break if limit_reached?(limit, index)
@@ -87,19 +85,6 @@ module Bulkrax
       status_info(e)
     end
 
-    def collections
-      records.map { |r| r[related_parents_parsed_mapping].split(/\s*[;|]\s*/) if r[related_parents_parsed_mapping].present? }.flatten.compact.uniq
-    end
-
-    def collections_total
-      collections.size
-    end
-
-    # TODO: change to differentiate between collection and work records when adding ability to import collection metadata
-    def works_total
-      total
-    end
-
     def total
       @total = importer.parser_fields['total'] || 0 if importer?
 
@@ -112,18 +97,6 @@ module Bulkrax
       @total = 0
     end
 
-    def extra_filters
-      output = ""
-      if importerexporter.start_date.present?
-        start_dt = importerexporter.start_date.to_datetime.strftime('%FT%TZ')
-        finish_dt = importerexporter.finish_date.present? ? importerexporter.finish_date.to_datetime.end_of_day.strftime('%FT%TZ') : "NOW"
-        output += " AND system_modified_dtsi:[#{start_dt} TO #{finish_dt}]"
-      end
-      output += importerexporter.work_visibility.present? ? " AND visibility_ssi:#{importerexporter.work_visibility}" : ""
-      output += importerexporter.workflow_status.present? ? " AND workflow_state_name_ssim:#{importerexporter.workflow_status}" : ""
-      output
-    end
-
     def current_record_ids
       @work_ids = []
       @collection_ids = []
@@ -140,78 +113,39 @@ module Bulkrax
       when 'importer'
         set_ids_for_exporting_from_importer
       end
-      @work_ids + @collection_ids + @file_set_ids
-    end
 
-    # Set the following instance variables: @work_ids, @collection_ids, @file_set_ids
-    # @see #current_record_ids
-    def set_ids_for_exporting_from_importer
-      entry_ids = Importer.find(importerexporter.export_source).entries.pluck(:id)
-      complete_statuses = Status.latest_by_statusable
-                                .includes(:statusable)
-                                .where('bulkrax_statuses.statusable_id IN (?) AND bulkrax_statuses.statusable_type = ? AND status_message = ?', entry_ids, 'Bulkrax::Entry', 'Complete')
-
-      complete_entry_identifiers = complete_statuses.map { |s| s.statusable&.identifier&.gsub(':', '\:') }
-      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.post(
-          extra_filters.to_s,
-          fq: [
-            %(#{::Solrizer.solr_name(work_identifier)}:("#{complete_entry_identifiers.join('" OR "')}")),
-            "has_model_ssim:(#{models_to_search.join(' OR ')})"
-          ],
-          fl: 'id',
-          rows: 2_000_000_000
-        )['response']['docs'].map { |obj| obj['id'] })
-      end
+      find_child_file_sets(@work_ids) if importerexporter.export_from == 'collection' || importerexporter.export_from == 'worktype'
+
+      @work_ids + @collection_ids + @file_set_ids
     end
 
     # export methods
 
-    def create_new_entries
-      current_record_ids.each_with_index do |id, index|
-        break if limit_reached?(limit, index)
-
-        this_entry_class = if @collection_ids.include?(id)
-                             collection_entry_class
-                           elsif @file_set_ids.include?(id)
-                             file_set_entry_class
-                           else
-                             entry_class
-                           end
-        new_entry = find_or_create_entry(this_entry_class, id, 'Bulkrax::Exporter')
-
-        begin
-          entry = ExportWorkJob.perform_now(new_entry.id, current_run.id)
-        rescue => e
-          Rails.logger.info("#{e.message} was detected during export")
-        end
-
-        self.headers |= entry.parsed_metadata.keys if entry
-      end
-    end
-    alias create_from_collection create_new_entries
-    alias create_from_importer create_new_entries
-    alias create_from_worktype create_new_entries
-    alias create_from_all create_new_entries
-
     # rubocop:disable Metrics/MethodLength, Metrics/AbcSize
     def write_files
       require 'open-uri'
       require 'socket'
+
+      folder_count = 1
+      records_in_folder = 0
+
       importerexporter.entries.where(identifier: current_record_ids)[0..limit || total].each do |entry|
-        work = ActiveFedora::Base.find(entry.identifier)
-        next unless Hyrax.config.curation_concerns.include?(work.class)
-        bag = BagIt::Bag.new setup_bagit_folder(entry.identifier)
+        record = ActiveFedora::Base.find(entry.identifier)
+        next unless Hyrax.config.curation_concerns.include?(record.class)
+
         bag_entries = [entry]
+        file_set_entries = Bulkrax::CsvFileSetEntry.where(importerexporter_id: importerexporter.id).where("parsed_metadata LIKE '%#{record.id}%'")
+        file_set_entries.each { |fse| bag_entries << fse }
 
-        work.file_sets.each do |fs|
-          if @file_set_ids.present?
-            file_set_entry = Bulkrax::CsvFileSetEntry.where("parsed_metadata LIKE '%#{fs.id}%'").first
-            bag_entries << file_set_entry unless file_set_entry.nil?
-          end
+        records_in_folder += bag_entries.count
+        if records_in_folder > records_split_count
+          folder_count += 1
+          records_in_folder = bag_entries.count
+        end
+
+        bag ||= BagIt::Bag.new setup_bagit_folder(folder_count, entry.identifier)
 
+        record.file_sets.each do |fs|
           file_name = filename(fs)
           next if file_name.blank?
           io = open(fs.original_file.uri)
@@ -226,17 +160,21 @@ module Bulkrax
           end
         end
 
-        CSV.open(setup_csv_metadata_export_file(entry.identifier), "w", headers: export_headers, write_headers: true) do |csv|
+        CSV.open(setup_csv_metadata_export_file(folder_count, entry.identifier), "w", headers: export_headers, write_headers: true) do |csv|
           bag_entries.each { |csv_entry| csv << csv_entry.parsed_metadata }
         end
-        write_triples(entry)
+
+        write_triples(folder_count, entry)
         bag.manifest!(algo: 'sha256')
       end
     end
     # rubocop:enable Metrics/MethodLength, Metrics/AbcSize
 
-    def setup_csv_metadata_export_file(id)
-      File.join(importerexporter.exporter_export_path, id, 'metadata.csv')
+    def setup_csv_metadata_export_file(folder_count, id)
+      path = File.join(importerexporter.exporter_export_path, folder_count.to_s)
+      FileUtils.mkdir_p(path) unless File.exist?(path)
+
+      File.join(path, id, 'metadata.csv')
     end
 
     def key_allowed(key)
@@ -245,66 +183,31 @@ module Bulkrax
         key != source_identifier.to_s
     end
 
-    # All possible column names
-    def export_headers
-      headers = sort_headers(self.headers)
-
-      # we don't want access_control_id exported and we want file at the end
-      headers.delete('access_control_id') if headers.include?('access_control_id')
+    def setup_triple_metadata_export_file(folder_count, id)
+      path = File.join(importerexporter.exporter_export_path, folder_count.to_s)
+      FileUtils.mkdir_p(path) unless File.exist?(path)
 
-      # add the headers below at the beginning or end to maintain the preexisting export behavior
-      headers.prepend('model')
-      headers.prepend(source_identifier.to_s)
-      headers.prepend('id')
-
-      headers.uniq
+      File.join(path, id, 'metadata.nt')
     end
 
-    def object_names
-      return @object_names if @object_names
-
-      @object_names = mapping.values.map { |value| value['object'] }
-      @object_names.uniq!.delete(nil)
-
-      @object_names
-    end
-
-    def sort_headers(headers)
-      # converting headers like creator_name_1 to creator_1_name so they get sorted by numerical order
-      # while keeping objects grouped together
-      headers.sort_by do |item|
-        number = item.match(/\d+/)&.[](0) || 0.to_s
-        sort_number = number.rjust(4, "0")
-        object_prefix = object_names.detect { |o| item.match(/^#{o}/) } || item
-        remainder = item.gsub(/^#{object_prefix}_/, '').gsub(/_#{number}/, '')
-        "#{object_prefix}_#{sort_number}_#{remainder}"
-      end
-    end
+    def setup_bagit_folder(folder_count, id)
+      path = File.join(importerexporter.exporter_export_path, folder_count.to_s)
+      FileUtils.mkdir_p(path) unless File.exist?(path)
 
-    def setup_triple_metadata_export_file(id)
-      File.join(importerexporter.exporter_export_path, id, 'metadata.nt')
+      File.join(path, id)
     end
 
-    def setup_bagit_folder(id)
-      File.join(importerexporter.exporter_export_path, id)
-    end
-
-    def write_triples(e)
+    def write_triples(folder_count, e)
       sd = SolrDocument.find(e.identifier)
       return if sd.nil?
 
       req = ActionDispatch::Request.new({ 'HTTP_HOST' => Socket.gethostname })
       rdf = Hyrax::GraphExporter.new(sd, req).fetch.dump(:ntriples)
-      File.open(setup_triple_metadata_export_file(e.identifier), "w") do |triples|
+      File.open(setup_triple_metadata_export_file(folder_count, e.identifier), "w") do |triples|
         triples.write(rdf)
       end
     end
 
-    def required_elements?(keys)
-      return if keys.blank?
-      !required_elements.map { |el| keys.map(&:to_s).include?(el) }.include?(false)
-    end
-
     # @todo - investigate getting directory structure
     # @todo - investigate using perform_later, and having the importer check for
     #   DownloadCloudFileJob before it starts
@@ -355,5 +258,11 @@ module Bulkrax
       return nil unless bag.valid?
       bag
     end
+
+    # use the version of this method from the application parser instead
+    def real_import_file_path
+      return importer_unzip_path if file? && zip?
+      parser_fields['import_file_path']
+    end
   end
 end

data/app/parsers/bulkrax/csv_parser.rb

@@ -4,6 +4,7 @@ require 'csv'
 module Bulkrax
   class CsvParser < ApplicationParser # rubocop:disable Metrics/ClassLength
     include ErroredEntries
+    include ExportBehavior
     attr_writer :collections, :file_sets, :works
 
     def self.export_supported?
@@ -207,6 +208,13 @@ module Bulkrax
       @work_ids + @collection_ids + @file_set_ids
     end
 
+    # find the related file set ids so entries can be made for export
+    def find_child_file_sets(work_ids)
+      work_ids.each do |id|
+        ActiveFedora::Base.find(id).file_set_ids.each { |fs_id| @file_set_ids << fs_id }
+      end
+    end
+
     # Set the following instance variables: @work_ids, @collection_ids, @file_set_ids
     # @see #current_record_ids
     def set_ids_for_exporting_from_importer
@@ -272,8 +280,8 @@ module Bulkrax
       CsvFileSetEntry
     end
 
-    # See https://stackoverflow.com/questions/2650517/count-the-number-of-lines-in-a-file-without-reading-entire-file-into-memory
-    #   Changed to grep as wc -l counts blank lines, and ignores the final unescaped line (which may or may not contain data)
+    # TODO: figure out why using the version of this method that's in the bagit parser
+    # breaks specs for the "if importer?" line
     def total
       @total = importer.parser_fields['total'] || 0 if importer?
       @total = limit || current_record_ids.count if exporter?
@@ -283,6 +291,10 @@ module Bulkrax
       @total = 0
     end
 
+    def records_split_count
+      1000
+    end
+
     # @todo - investigate getting directory structure
     # @todo - investigate using perform_later, and having the importer check for
     #   DownloadCloudFileJob before it starts
@@ -307,9 +319,37 @@ module Bulkrax
     # export methods
 
     def write_files
-      CSV.open(setup_export_file, "w", headers: export_headers, write_headers: true) do |csv|
-        importerexporter.entries.where(identifier: current_record_ids)[0..limit || total].each do |e|
-          csv << e.parsed_metadata
+      require 'open-uri'
+      folder_count = 0
+
+      importerexporter.entries.where(identifier: current_record_ids)[0..limit || total].in_groups_of(records_split_count, false) do |group|
+        folder_count += 1
+
+        CSV.open(setup_export_file(folder_count), "w", headers: export_headers, write_headers: true) do |csv|
+          group.each do |entry|
+            csv << entry.parsed_metadata
+            next if importerexporter.metadata_only? || entry.type == 'Bulkrax::CsvCollectionEntry'
+
+            store_files(entry.identifier, folder_count.to_s)
+          end
+        end
+      end
+    end
+
+    def store_files(identifier, folder_count)
+      record = ActiveFedora::Base.find(identifier)
+      file_sets = record.file_set? ? Array.wrap(record) : record.file_sets
+      file_sets << record.thumbnail if exporter.include_thumbnails && record.thumbnail.present? && record.work?
+      file_sets.each do |fs|
+        path = File.join(exporter_export_path, folder_count, 'files')
+        FileUtils.mkdir_p(path) unless File.exist? path
+        file = filename(fs)
+        io = open(fs.original_file.uri)
+        next if file.blank?
+
+        File.open(File.join(path, file), 'wb') do |f|
+          f.write(io.read)
+          f.close
         end
       end
     end
@@ -356,8 +396,11 @@ module Bulkrax
     end
 
     # in the parser as it is specific to the format
-    def setup_export_file
-      File.join(importerexporter.exporter_export_path, "export_#{importerexporter.export_source}_from_#{importerexporter.export_from}.csv")
+    def setup_export_file(folder_count)
+      path = File.join(importerexporter.exporter_export_path, folder_count.to_s)
+      FileUtils.mkdir_p(path) unless File.exist?(path)
+
+      File.join(path, "export_#{importerexporter.export_source}_from_#{importerexporter.export_from}_#{folder_count}.csv")
     end
 
     # Retrieve file paths for [:file] mapping in records
@@ -382,10 +425,11 @@ module Bulkrax
     end
 
     # Retrieve the path where we expect to find the files
-    def path_to_files
+    def path_to_files(**args)
+      filename = args.fetch(:filename, '')
+
       @path_to_files ||= File.join(
-        zip? ? importer_unzip_path : File.dirname(import_file_path),
-        'files'
+        zip? ? importer_unzip_path : File.dirname(import_file_path), 'files', filename
       )
     end
 

data/app/views/bulkrax/exporters/_downloads.html.erb

@@ -0,0 +1,8 @@
+<%= form.select :exporter_export_zip_files,
+  exporter.sort_zip_files(form.object.exporter_export_zip_files.flatten),
+  {},
+  {
+    class: 'btn btn-default form-control',
+    style: 'width: 200px'
+  }
+%>

data/app/views/bulkrax/exporters/_form.html.erb

@@ -29,6 +29,7 @@
 
   <%= form.input :export_source_importer,
     label: t('bulkrax.exporter.labels.importer'),
+    # required: true,
     prompt: 'Select from the list',
     label_html: { class: 'importer export-source-option hidden' },
     input_html: { class: 'importer export-source-option hidden' },
@@ -37,6 +38,7 @@
   <%= form.input :export_source_collection,
     prompt: 'Start typing ...',
     label: t('bulkrax.exporter.labels.collection'),
+    # required: true,
     placeholder: @collection&.title&.first,
     label_html: { class: 'collection export-source-option hidden' },
     input_html: {
@@ -50,6 +52,7 @@
 
   <%= form.input :export_source_worktype,
     label: t('bulkrax.exporter.labels.worktype'),
+    # required: true,
     prompt: 'Select from the list',
     label_html: { class: 'worktype export-source-option hidden' },
     input_html: { class: 'worktype export-source-option hidden' },

data/app/views/bulkrax/exporters/index.html.erb

@@ -21,7 +21,7 @@
               <th scope="col">Name</th>
               <th scope="col">Status</th>
               <th scope="col">Date Exported</th>
-              <th scope="col"></th>
+              <th scope="col">Downloadable Files</th>
               <th scope="col"></th>
               <th scope="col"></th>
               <th scope="col"></th>
@@ -35,7 +35,10 @@
                 <td><%= exporter.created_at %></td>
                 <td>
                   <% if File.exist?(exporter.exporter_export_zip_path) %>
-                    <%= link_to raw('<span class="glyphicon glyphicon-download"></span>'), exporter_download_path(exporter) %>
+                    <%= simple_form_for(exporter, method: :get, url: exporter_download_path(exporter)) do |form| %>
+                      <%= render 'downloads', exporter: exporter, form: form %>
+                      <%= form.button :submit, value: 'Download', data: { disable_with: false } %>
+                    <% end %>
                   <% end%>
                 </td>
                 <td><%= link_to raw('<span class="glyphicon glyphicon-info-sign"></span>'), exporter_path(exporter) %></td>

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

@@ -8,10 +8,11 @@
   <div class='panel-body'>
 
     <% if File.exist?(@exporter.exporter_export_zip_path) %>
-      <p class='bulkrax-p-align'>
+      <%= simple_form_for @exporter, method: :get, url: exporter_download_path(@exporter), html: { class: 'form-inline bulkrax-p-align' } do |form| %>
         <strong>Download:</strong>
-        <%= link_to raw('<span class="glyphicon glyphicon-download"></span>'), exporter_download_path(@exporter) %>
-      </p>
+        <%= render 'downloads', exporter: @exporter, form: form %>
+        <%= form.button :submit, value: 'Download', data: { disable_with: false } %>
+      <% end %>
     <% end %>
 
     <p class='bulkrax-p-align'>
@@ -135,10 +136,6 @@
       <%= page_entries_info(@work_entries) %><br>
       <%= paginate(@work_entries, param_name: :work_entries_page) %>
       <br>
-      <% if File.exist?(@exporter.exporter_export_zip_path) %>
-        <%= link_to 'Download', exporter_download_path(@exporter) %>
-        |
-      <% end %>
       <%= link_to 'Edit', edit_exporter_path(@exporter) %>
       |
       <%= link_to 'Back', exporters_path %>

data/lib/bulkrax/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Bulkrax
-  VERSION = '3.4.0'
+  VERSION = '4.0.0'
 end

data/lib/tasks/bulkrax_tasks.rake

@@ -1,6 +1,30 @@
 # frozen_string_literal: true
 
-# desc "Explaining what the task does"
-# task :bulkrax do
-#   # Task goes here
-# end
+namespace :bulkrax do
+  desc "Remove old exported zips and create new ones with the new file structure"
+  task rerun_all_exporters: :environment do
+    if defined?(::Hyku)
+      Account.find_each do |account|
+        puts "=============== updating #{account.name} ============"
+        next if account.name == "search"
+        switch!(account)
+
+        rerun_exporters_and_delete_zips
+
+        puts "=============== finished updating #{account.name} ============"
+      end
+    else
+      rerun_exporters_and_delete_zips
+    end
+  end
+
+  def rerun_exporters_and_delete_zips
+    begin
+      Bulkrax::Exporter.all.each { |e| Bulkrax::ExporterJob.perform_later(e.id) }
+    rescue => e
+      puts "(#{e.message})"
+    end
+
+    Dir["tmp/exports/**.zip"].each { |zip_path| FileUtils.rm_rf(zip_path) }
+  end
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: bulkrax
 version: !ruby/object:Gem::Version
-  version: 3.4.0
+  version: 4.0.0
 platform: ruby
 authors:
 - Rob Kaufman
-autorequire: 
+autorequire:
 bindir: bin
 cert_chain: []
-date: 2022-06-22 00:00:00.000000000 Z
+date: 2022-07-15 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rails
@@ -331,6 +331,7 @@ files:
 - app/views/bulkrax/entries/_parsed_metadata.html.erb
 - app/views/bulkrax/entries/_raw_metadata.html.erb
 - app/views/bulkrax/entries/show.html.erb
+- app/views/bulkrax/exporters/_downloads.html.erb
 - app/views/bulkrax/exporters/_form.html.erb
 - app/views/bulkrax/exporters/edit.html.erb
 - app/views/bulkrax/exporters/index.html.erb
@@ -404,7 +405,7 @@ homepage: https://github.com/samvera-labs/bulkrax
 licenses:
 - Apache-2.0
 metadata: {}
-post_install_message: 
+post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -419,8 +420,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.1.4
-signing_key: 
+rubygems_version: 3.0.3
+signing_key:
 specification_version: 4
 summary: Import and export tool for Hyrax and Hyku
 test_files: []