2979 - Metadata template and download should be consistent

app/controllers/concerns/export_service.rb

@@ -126,9 +126,11 @@ def export_collection_contributors_csv(out:, collection:, report_arguments:)
   end
 
   def export_work_metadata_csv(out:, collection:)
-    path = "work_metadata.csv"
+    path = 'work_metadata.csv'
     out.put_next_entry(path)
-    out.write(export_work_metadata_as_csv(collection))
+
+    result = Work::Metadata::ExportCsv.call(collection: collection, works: collection.works)
+    out.write(result.csv_string)
   end
 
   def export_subject_csv(out:, collection:, work:)
@@ -341,7 +343,6 @@ def export_html_full_pages(out:, page:, by_work:, original_filenames:)
       path = File.join("html_full_pages", "#{path_from_work(page.work, original_filenames)}_#{page.title}.html")
     end
 
-
     out.put_next_entry path
 
     page_view = xml_to_html(page.xml_text, true, false, page.work.collection)
@@ -402,107 +403,6 @@ def get_headings(collection, ids)
     @headings
   end
 
-
-  def export_work_metadata_as_csv(collection)
-    csv_string = CSV.generate(:force_quotes => true) do |csv|
-      static_headers = [
-        'Title',
-        'Collection',
-        'Document Sets',
-        'Uploaded Filename',
-        'FromThePage ID',
-        'FromThePage Slug',
-        'FromThePage URL',
-        'Identifier',
-        'Originating Manifest ID',
-        'Creation Date',
-        'Total Pages',
-        'Pages Transcribed',
-        'Pages Corrected',
-        'Pages Indexed',
-        'Pages Translated',
-        'Pages Needing Review',
-        'Pages Marked Blank',
-        'Contributors',
-        'Contributors Name',
-        'work_id'
-      ]
-
-      raw_metadata_strings = collection.works.pluck(:original_metadata)
-      metadata_headers = raw_metadata_strings.map{|raw| raw.nil? ? [] : JSON.parse(raw).map{|element| element["label"] } }.flatten.uniq
-      # append the headers for described metadata, read from the metadata_field configuration for the project
-      static_description_headers = ['Description Status', 'Described By']
-      described_headers = collection.metadata_fields.map {|field| field.label}
-
-      csv << static_headers + metadata_headers + static_description_headers + described_headers
-
-      collection.works.includes(:document_sets, :work_statistic, :sc_manifest).reorder(:id).each do |work|
-
-        work_users = work.deeds.map{ |d| "#{d.user.display_name}<#{d.user.email}>".gsub('|', '//') }.uniq.join('|')
-        contributors_real_names = work.deeds.map{ |d| d.user.real_name }.uniq.join(' | ')
-        row = [
-          work.title,
-          work.collection.title,
-          work.document_sets.map{|ds| ds.title}. join('|'),
-          work.uploaded_filename,
-          work.id,
-          work.slug,
-          collection_read_work_url(collection.owner, collection, work),
-          work.identifier,
-          work.sc_manifest.nil? ? '' : work.sc_manifest.at_id,
-          work.created_on,
-          work.work_statistic.total_pages,
-          work.work_statistic.transcribed_pages,
-          work.work_statistic.corrected_pages,
-          work.work_statistic.annotated_pages,
-          work.work_statistic.translated_pages,
-          work.work_statistic.needs_review,
-          work.work_statistic.blank_pages,
-          work_users,
-          contributors_real_names,
-          work.id
-
-        ]
-
-        unless work.original_metadata.blank?
-          metadata = {}
-          JSON.parse(work.original_metadata).each {|e| metadata[e['label']] = e['value'] }
-
-          metadata_headers.each do |header|
-            # look up the value for this index
-            row << metadata[header]
-          end
-        end
-
-        unless work.metadata_description.blank?
-          # description status
-          row << work.description_status
-          # described by
-          row << User.find(work.metadata_description_versions.pluck(:user_id)).map{|u| u.display_name}.join('; ')
-
-          metadata = JSON.parse(work.metadata_description)
-          # we rely on a consistent order of fields returned by collection.metadata_fields to prevent scrambling columns
-          collection.metadata_fields.each do |field|
-            element = metadata.detect{|candidate| candidate['transcription_field_id'] == field.id}
-            if element
-              value = element['value']
-              if value.is_a? Array
-                value = value.join("; ")
-              end
-              row << value
-            else
-              row << nil
-            end
-          end
-        end
-
-        csv << row
-      end
-    end
-
-    csv_string
-  end
-
   def export_tables_as_csv(table_obj)
     if table_obj.is_a?(Collection)
       collection = table_obj

app/controllers/export_controller.rb

@@ -89,9 +89,12 @@ def subject_index_csv
   end
 
   def work_metadata_csv
+    filename = params[:filename] ? "#{params[:filename]}.csv" : "fromthepage_work_metadata_export_#{@collection.id}_#{Time.now.utc.iso8601}.csv"
+    result = Work::Metadata::ExportCsv.call(collection: @collection, works: @collection.works)
+
     send_data(
-      export_work_metadata_as_csv(@collection),
-      filename: "fromthepage_work_metadata_export_#{@collection.id}_#{Time.now.utc.iso8601}.csv",
+      result.csv_string,
+      filename: filename,
       type: 'application/csv'
     )
     cookies['download_finished'] = 'true'

app/controllers/metadata_controller.rb

@@ -3,28 +3,24 @@ class MetadataController < ApplicationController
 
   def example
     collection = Collection.find(params[:id])
-    example = Metadata.create_example(collection)
-    send_data example, filename: "example.csv"
+    works = collection.works.limit(3)
+    result = Work::Metadata::ExportCsv.call(collection: collection, works: works)
+
+    send_data result.csv_string, filename: 'example.csv'
+  end
+
+  def upload
+    # Modal upload
   end
 
   def create
-    metadata_file = params[:metadata]['file'].tempfile
-
-    collection = Collection.find(params[:metadata][:collection_id])
-    metadata = Metadata.new(metadata_file: metadata_file, collection: collection)
-    result = metadata.process_csv
-    rows = result[:content]
-    row_errors = result[:errors].count
-    link = helpers.link_to 'link', collection_metadata_csv_error_path
-
-    if row_errors > 0
-      feedback = "Your upload has finished processing. #{rows} works were updated successfully; #{row_errors} rows encountered errors. Download the error file here: #{link}"
-    else
-      feedback = "Your upload has finished processing. #{rows} works were updated successfully."
-    end
+    metadata_file_path = params[:metadata]['file'].tempfile.path
+    collection_id = params[:metadata][:collection_id]
+    Work::Metadata::ImportCsvJob.perform_later(metadata_file_path, collection_id, current_user.id)
 
-    flash[:alert] = feedback
+    collection = Collection.find(collection_id)
 
+    flash[:alert] = t('.is_processing')
     ajax_redirect_to edit_look_collection_path(collection.owner, collection)
   end
 
@@ -46,9 +42,4 @@ def refresh
     flash[:notice] = t('.is_processing')
     ajax_redirect_to edit_look_collection_path(collection.owner, collection)
   end
-
-  def csv_error
-    csv_string = Metadata.retrieve_error
-    send_data csv_string, filename: "error.csv"
-  end
 end

app/interactors/work/metadata/export_csv.rb

@@ -0,0 +1,126 @@
+require 'csv'
+
+class Work::Metadata::ExportCsv
+  STATIC_HEADERS = [
+    'FromThePage Title',
+    '*Collection*',
+    '*Document Sets*',
+    '*Uploaded Filename*',
+    '*FromThePage ID*',
+    '*FromThePage Slug*',
+    '*FromThePage URL*',
+    'FromThePage Description',
+    'Identifier',
+    '*Originating Manifest ID*',
+    '*Creation Date*',
+    '*Total Pages*',
+    '*Pages Transcribed*',
+    '*Pages Corrected*',
+    '*Pages Indexed*',
+    '*Pages Translated*',
+    '*Pages Needing Review*',
+    '*Pages Marked Blank*',
+    '*Contributors*',
+    '*Contributors Name*'
+  ].freeze
+
+  STATIC_DESCRIPTION_HEADERS = [
+    '*Description Status*',
+    '*Described By*'
+  ].freeze
+
+  include Interactor
+  include Rails.application.routes.url_helpers
+
+  def initialize(collection:, works:)
+    @collection = collection
+    @works = works
+
+    super
+  end
+
+  def call
+    csv_string = CSV.generate(force_quotes: true) do |csv|
+      works_scope = @works.includes(
+                             :document_sets,
+                             :work_statistic,
+                             :sc_manifest,
+                             :deeds,
+                             { metadata_description_versions: :user }
+                           )
+                          .reorder(:id)
+
+      raw_metadata_strings = works_scope.pluck(:original_metadata)
+      metadata_headers = raw_metadata_strings
+                         .compact
+                         .flat_map { |raw| JSON.parse(raw).map { |element| element['label'] } }
+                         .uniq
+
+      described_headers = @collection.metadata_fields.map(&:label)
+      csv << STATIC_HEADERS + metadata_headers + STATIC_DESCRIPTION_HEADERS + described_headers
+
+      works_scope.each do |work|
+        work_users = work.deeds.map { |d| "#{d.user.display_name}<#{d.user.email}>".gsub('|', '//') }.uniq.join('|')
+        contributors_real_names = work.deeds.map { |d| d.user.real_name }.uniq.join(' | ')
+        row = [
+          work.title,
+          work.collection.title,
+          work.document_sets.map(&:title).join('|'),
+          work.uploaded_filename,
+          work.id,
+          work.slug,
+          collection_read_work_url(@collection.owner, @collection, work),
+          work.description,
+          work.identifier,
+          work.sc_manifest.nil? ? '' : work.sc_manifest.at_id,
+          work.created_on,
+          work.work_statistic.total_pages,
+          work.work_statistic.transcribed_pages,
+          work.work_statistic.corrected_pages,
+          work.work_statistic.annotated_pages,
+          work.work_statistic.translated_pages,
+          work.work_statistic.needs_review,
+          work.work_statistic.blank_pages,
+          work_users,
+          contributors_real_names
+        ]
+
+        if work.original_metadata.present?
+          metadata = {}
+          JSON.parse(work.original_metadata).each { |e| metadata[e['label']] = e['value'] }
+
+          metadata_headers.each do |header|
+            # look up the value for this index
+            row << metadata[header]
+          end
+        end
+
+        if work.metadata_description.present?
+          # description status
+          row << work.description_status
+          # described by
+          row << work.metadata_description_versions.flat_map(&:user).map(&:display_name).join('; ')
+
+          metadata = JSON.parse(work.metadata_description)
+
+          # we rely on a consistent order of fields returned by collection.metadata_fields to prevent scrambling columns
+          @collection.metadata_fields.each do |field|
+            element = metadata.detect { |candidate| candidate['transcription_field_id'] == field.id }
+            if element
+              value = element['value']
+              value = value.join('; ') if value.is_a? Array
+              row << value
+            else
+              row << nil
+            end
+          end
+        end
+
+        csv << row
+      end
+    end
+
+    context.csv_string = csv_string
+    context
+  end
+end