Class: Bulkrax::CsvParser

Inherits:
ApplicationParser show all
Includes:
ErroredEntries, ExportBehavior
Defined in:
app/parsers/bulkrax/csv_parser.rb

Overview

rubocop:disable Metrics/ClassLength

Direct Known Subclasses

BagitParser

Instance Attribute Summary collapse

Attributes inherited from ApplicationParser

#headers, #importerexporter

Class Method Summary collapse

Instance Method Summary collapse

Methods included from ExportBehavior

#build_export_metadata, #build_for_exporter, #filename, #hyrax_record

Methods included from ErroredEntries

#build_errored_entry_row, #setup_errored_entries_file, #write_errored_entries_file

Methods inherited from ApplicationParser

#base_path, #calculate_type_delay, #create_collections, #create_entry_and_job, #create_file_sets, #create_objects, #create_relationships, #create_works, #exporter?, #find_or_create_entry, #generated_metadata_mapping, #get_field_mapping_hash_for, #import_file_path, import_supported?, #importer?, #initialize, #invalid_record, #limit_reached?, #model_field_mappings, #new_entry, parser_fields, #path_for_import, #perform_method, #rebuild_entries, #rebuild_entry_query, #record, #record_deleted?, #record_has_source_identifier, #record_raw_metadata, #record_remove_and_rerun?, #related_children_parsed_mapping, #related_children_raw_mapping, #related_parents_parsed_mapping, #related_parents_raw_mapping, #required_elements, #source_identifier, #untar, #unzip, #visibility, #work_entry_class, #work_identifier, #work_identifier_search_field, #write, #write_import_file, #zip

Constructor Details

This class inherits a constructor from Bulkrax::ApplicationParser

Instance Attribute Details

#collections ⇒ Object

rubocop:enabled Metrics/AbcSize



59
60
61
62
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 59

def collections
  build_records if @collections.nil?
  @collections
end

#file_sets ⇒ Object 



69
70
71
72
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 69

def file_sets
  build_records if @file_sets.nil?
  @file_sets
end

#works ⇒ Object 



64
65
66
67
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 64

def works
  build_records if @works.nil?
  @works
end

Class Method Details

.export_supported? ⇒ Boolean

Returns:

  • (Boolean) 


9
10
11
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 9

def self.export_supported?
  true
end

Instance Method Details

#build_records ⇒ Object

rubocop:disable Metrics/AbcSize



26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 26

def build_records
  @collections = []
  @works = []
  @file_sets = []

  if model_field_mappings.map { |mfm| mfm.to_sym.in?(records.first.keys) }.any?
    records.map do |r|
      model_field_mappings.map(&:to_sym).each do |model_mapping|
        next unless r.key?(model_mapping)

        model = r[model_mapping].nil? ? "" : r[model_mapping].strip
        # TODO: Eventually this should be refactored to us Hyrax.config.collection_model
        #       We aren't right now because so many Bulkrax users are in between Fedora and Valkyrie
        if model.casecmp('collection').zero? || model.casecmp('collectionresource').zero?
          @collections << r
        elsif model.casecmp('fileset').zero?
          @file_sets << r
        else
          @works << r
        end
      end
    end
    @collections = @collections.flatten.compact.uniq
    @file_sets = @file_sets.flatten.compact.uniq
    @works = @works.flatten.compact.uniq
  else # if no model is specified, assume all records are works
    @works = records.flatten.compact.uniq
  end

  true
end

#collection_entry_class ⇒ Object 



160
161
162
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 160

def collection_entry_class
  CsvCollectionEntry
end

#collections_total ⇒ Object 



74
75
76
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 74

def collections_total
  collections.size
end

#create_new_entries ⇒ Object Also known as: create_from_collection, create_from_importer, create_from_worktype, create_from_all 



138
139
140
141
142
143
144
145
146
147
148
149
150
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 138

def create_new_entries
  # NOTE: The each method enforces the limit, as it can best optimize the underlying queries.
  current_records_for_export.each do |id, entry_class|
    new_entry = find_or_create_entry(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

#current_records_for_export ⇒ Object 



131
132
133
134
135
136
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 131

def current_records_for_export
  @current_records_for_export ||= Bulkrax::ParserExportRecordSet.for(
    parser: self,
    export_from: importerexporter.export_from
  )
end

#entry_class ⇒ Object 



156
157
158
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 156

def entry_class
  CsvEntry
end

#export_headers ⇒ Object

All possible column names



271
272
273
274
275
276
277
278
279
280
281
282
283
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 271

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')

  # 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
end

#export_key_allowed(key) ⇒ Object 



265
266
267
268
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 265

def export_key_allowed(key)
  new_entry(entry_class, 'Bulkrax::Exporter').field_supported?(key) &&
    key != source_identifier.to_s
end

#file_paths ⇒ Object

Retrieve file paths for [:file] mapping in records

and check all listed files exist.

Raises:

  • (StandardError) 


333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 333

def file_paths
  raise StandardError, 'No records were found' if records.blank?
  return [] if importerexporter.metadata_only?

  @file_paths ||= records.map do |r|
    file_mapping = Bulkrax.field_mappings.dig(self.class.to_s, 'file', :from)&.first&.to_sym || :file
    next if r[file_mapping].blank?

    r[file_mapping].split(Bulkrax.multi_value_element_split_on).map do |f|
      file = File.join(path_to_files, f.tr(' ', '_'))
      if File.exist?(file) # rubocop:disable Style/GuardClause
        file
      else
        raise "File #{file} does not exist"
      end
    end
  end.flatten.compact.uniq
end

#file_set_entry_class ⇒ Object 



164
165
166
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 164

def file_set_entry_class
  CsvFileSetEntry
end

#file_sets_total ⇒ Object 



82
83
84
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 82

def file_sets_total
  file_sets.size
end

#import_fields ⇒ Object

We could use CsvEntry#fields_from_data(data) but that would mean re-reading the data



87
88
89
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 87

def import_fields
  @import_fields ||= records.inject(:merge).keys.compact.uniq
end

#missing_elements(record) ⇒ Object 



95
96
97
98
99
100
101
102
103
104
105
106
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 95

def missing_elements(record)
  keys_from_record = keys_without_numbers(record.reject { |_, v| v.blank? }.keys.compact.uniq.map(&:to_s))
  keys = []
  # Because we're persisting the mapping in the database, these are likely string keys.
  # However, there's no guarantee.  So, we need to ensure that by running stringify.
  importerexporter.mapping.stringify_keys.map do |k, v|
    Array.wrap(v['from']).each do |vf|
      keys << k if keys_from_record.include?(vf)
    end
  end
  required_elements.map(&:to_s) - keys.uniq.map(&:to_s)
end

#object_names ⇒ Object 



285
286
287
288
289
290
291
292
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 285

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

#path_to_files(**args) ⇒ Object

Retrieve the path where we expect to find the files



353
354
355
356
357
358
359
360
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 353

def path_to_files(**args)
  filename = args.fetch(:filename, '')

  return @path_to_files if @path_to_files.present? && filename.blank?
  @path_to_files = File.join(
      zip? ? importer_unzip_path : File.dirname(import_file_path), 'files', filename
    )
end

#records(_opts = {}) ⇒ Object 



13
14
15
16
17
18
19
20
21
22
23
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 13

def records(_opts = {})
  return @records if @records.present?

  file_for_import = only_updates ? parser_fields['partial_import_file_path'] : import_file_path
  # data for entry does not need source_identifier for csv, because csvs are read sequentially and mapped after raw data is read.
  csv_data = entry_class.read_data(file_for_import)
  importer.parser_fields['total'] = csv_data.count
  importer.save

  @records = csv_data.map { |record_data| entry_class.data_for_entry(record_data, nil, self) }
end

#records_split_count ⇒ Object 



189
190
191
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 189

def records_split_count
  1000
end

#required_elements?(record) ⇒ Boolean

Returns:

  • (Boolean) 


91
92
93
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 91

def required_elements?(record)
  missing_elements(record).blank?
end

#retrieve_cloud_files(files, importer) ⇒ Object

TODO:

  • investigate getting directory structure

TODO:

  • investigate using perform_later, and having the importer check for

DownloadCloudFileJob before it starts



196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 196

def retrieve_cloud_files(files, importer)
  files_path = File.join(path_for_import, 'files')
  FileUtils.mkdir_p(files_path) unless File.exist?(files_path)
  target_files = []
  files.each_pair do |_key, file|
    # fixes bug where auth headers do not get attached properly
    if file['auth_header'].present?
      file['headers'] ||= {}
      file['headers'].merge!(file['auth_header'])
    end
    # this only works for uniquely named files
    target_file = File.join(files_path, file['file_name'].tr(' ', '_'))
    target_files << target_file
    # Now because we want the files in place before the importer runs
    # Problematic for a large upload
    Bulkrax::DownloadCloudFileJob.perform_later(file, target_file)
  end
  importer[:parser_fields]['original_file_paths'] = target_files
  return nil
end

#setup_export_file(folder_count) ⇒ Object

in the parser as it is specific to the format



324
325
326
327
328
329
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 324

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

#sort_entries(entries) ⇒ Object 



294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 294

def sort_entries(entries)
  # always export models in the same order: work, collection, file set
  #
  # TODO: This is a problem in that only these classes are compared.  Instead
  #       We should add a comparison operator to the classes.
  entries.sort_by do |entry|
    case entry.type
    when 'Bulkrax::CsvCollectionEntry'
      '1'
    when 'Bulkrax::CsvFileSetEntry'
      '2'
    else
      '0'
    end
  end
end

#sort_headers(headers) ⇒ Object 



311
312
313
314
315
316
317
318
319
320
321
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 311

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

#store_files(identifier, folder_count) ⇒ Object 



243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 243

def store_files(identifier, folder_count)
  record = Bulkrax.object_factory.find(identifier)
  return unless record

  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)
    next if file.blank? || fs.original_file.blank?

    io = open(fs.original_file.uri)
    File.open(File.join(path, file), 'wb') do |f|
      f.write(io.read)
      f.close
    end
  end
rescue Ldp::Gone
  return
end

#total ⇒ Object

TODO: figure out why using the version of this method that’s in the bagit parser breaks specs for the “if importer?” line



174
175
176
177
178
179
180
181
182
183
184
185
186
187
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 174

def total
  @total =
    if importer?
      importer.parser_fields['total'] || 0
    elsif exporter?
      limit.to_i.zero? ? current_records_for_export.count : limit.to_i
    else
      0
    end

  return @total
rescue StandardError
  @total = 0
end

#valid_entry_types ⇒ Object 



168
169
170
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 168

def valid_entry_types
  [collection_entry_class.to_s, file_set_entry_class.to_s, entry_class.to_s]
end

#valid_import? ⇒ Boolean

Returns:

  • (Boolean) 


108
109
110
111
112
113
114
115
116
117
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 108

def valid_import?
  compressed_record = records.flat_map(&:to_a).partition { |_, v| !v }.flatten(1).to_h
  error_alert = "Missing at least one required element, missing element(s) are: #{missing_elements(compressed_record).join(', ')}"
  raise StandardError, error_alert unless required_elements?(compressed_record)

  file_paths.is_a?(Array)
rescue StandardError => e
  set_status_info(e)
  false
end

#works_total ⇒ Object 



78
79
80
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 78

def works_total
  works.size
end

#write_files ⇒ Object

export methods



219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 219

def write_files
  require 'open-uri'
  folder_count = 0
  # TODO: This is not performant as well; unclear how to address, but lower priority as of
  #       <2023-02-21 Tue>.
  sorted_entries = sort_entries(importerexporter.entries.uniq(&:identifier))
                   .select { |e| valid_entry_types.include?(e.type) }

  group_size = limit.to_i.zero? ? total : limit.to_i
  sorted_entries[0..group_size].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
        # TODO: This is precarious when we have descendents of Bulkrax::CsvCollectionEntry
        next if importerexporter.metadata_only? || entry.type == 'Bulkrax::CsvCollectionEntry'

        store_files(entry.identifier, folder_count.to_s)
      end
    end
  end
end

#write_partial_import_file(file) ⇒ Object 



119
120
121
122
123
124
125
126
127
128
129
 
# File 'app/parsers/bulkrax/csv_parser.rb', line 119

def write_partial_import_file(file)
  import_filename = import_file_path.split('/').last
  partial_import_filename = "#{File.basename(import_filename, '.csv')}_corrected_entries.csv"

  path = File.join(path_for_import, partial_import_filename)
  FileUtils.mv(
    file.path,
    path
  )
  path
end