Batch sample import service

app/services/samples/batch_file_import_service.rb

@@ -0,0 +1,172 @@
+# frozen_string_literal: true
+
+require 'roo'
+
+module Samples
+  # Service used to batch create samples via a file
+  class BatchFileImportService < BaseService # rubocop:disable Metrics/ClassLength
+    SampleFileImportError = Class.new(StandardError)
+
+    def initialize(namespace, user = nil, blob_id = nil, params = {})
+      super(user, params)
+      @namespace = namespace
+      @file = ActiveStorage::Blob.find(blob_id)
+      @sample_name_column = params[:sample_name_column]
+      @project_puid_column = params[:project_puid_column]
+      @sample_description_column = params[:sample_description_column]
+      @spreadsheet = nil
+      @headers = nil
+      @temp_import_file = Tempfile.new
+    end
+
+    def execute
+      authorize! @namespace, to: :update_sample_metadata?
+
+      validate_sample_name_column
+
+      validate_file
+
+      perform_file_import
+    rescue Samples::BatchFileImportService::SampleFileImportError => e
+      @namespace.errors.add(:base, e.message)
+      {}
+    end
+
+    private
+
+    def validate_sample_name_column
+      return unless @sample_name_column.nil?
+
+      raise SampleFileImportError,
+            I18n.t('services.samples.metadata.import_file.empty_sample_id_column') # TODO: refactor out
+    end
+
+    def validate_file_extension
+      file_extension = File.extname(@file.filename.to_s).downcase
+
+      return file_extension if %w[.csv .tsv .xls .xlsx].include?(file_extension)
+
+      raise SampleFileImportError,
+            I18n.t('services.samples.metadata.import_file.invalid_file_extension') # TODO: refactor out
+    end
+
+    def validate_file_headers
+      duplicate_headers = @headers.find_all { |header| @headers.count(header) > 1 }.uniq
+      unless duplicate_headers.empty?
+        raise SampleFileImportError,
+              I18n.t('services.sammple.batch_import.duplicate_column_names')
+      end
+
+      unless @headers.include?(@sample_name_column)
+        raise SampleFileImportError,
+              I18n.t('services.samples.batch_import.missing_header',
+                     header_title: @sample_name_column, header_desc: 'sample name')
+      end
+
+      unless @headers.include?(@project_puid_column) # rubocop:disable Style/GuardClause
+        raise SampleFileImportError,
+              I18n.t('services.samples.batch_import.missing_header',
+                     header_title: @project_puid_column, header_desc: 'project puid')
+      end
+    end
+
+    def validate_file_rows
+      # Should have at least 2 rows
+      first_row = @spreadsheet.row(2)
+      return unless first_row.compact.empty?
+
+      raise SampleFileImportError,
+            I18n.t('services.samples.batch_import.missing_data_row')
+    end
+
+    def validate_file
+      if @file.nil?
+        raise SampleFileImportError,
+              I18n.t('services.samples.batch_import.empty_file')
+      end
+
+      extension = validate_file_extension
+      download_batch_import_file(extension)
+
+      @headers = @spreadsheet.row(1).compact
+      validate_file_headers
+
+      validate_file_rows
+    end
+
+    def download_batch_import_file(extension)
+      begin
+        @temp_import_file.binmode
+        @file.download do |chunk|
+          @temp_import_file.write(chunk)
+        end
+      ensure
+        @temp_import_file.close
+      end
+      @spreadsheet = if extension.eql? '.tsv'
+                       Roo::CSV.new(@temp_import_file, extension:, csv_options: { col_sep: "\t" })
+                     else
+                       Roo::Spreadsheet.open(@temp_import_file.path, extension:)
+                     end
+    end
+
+    def perform_file_import # rubocop:disable Metrics/AbcSize,Metrics/MethodLength
+      response = {}
+      parse_settings = @headers.zip(@headers).to_h
+
+      @spreadsheet.each_with_index(parse_settings) do |data, index|
+        next unless index.positive?
+
+        # TODO: handle metadata
+
+        sample_name = data[@sample_name_column]
+        project_puid = data[@project_puid_column]
+        description = data[@sample_description_column]
+
+        if sample_name.nil? || project_puid.nil?
+          response["index #{index}"] = {
+            path: ['sample'],
+            message: I18n.t('services.samples.batch_import.missing_field', index: index)
+          }
+          next
+        end
+
+        project = Namespaces::ProjectNamespace.find_by(puid: project_puid)&.project
+        unless project
+          response[sample_name] = {
+            path: ['project'],
+            message: I18n.t('services.samples.batch_import.project_puid_not_found', project_puid: project_puid)
+          }
+          next
+        end
+
+        response[sample_name] = process_sample_row(sample_name, project, description)
+        cleanup_files
+        response
+      end
+      response
+    end
+
+    def cleanup_files
+      # delete the blob and temporary file as we no longer require them
+      @file.purge
+      @temp_import_file.unlink
+    end
+
+    def process_sample_row(name, project, description)
+      sample_params = { name:, description: }
+      sample = Samples::CreateService.new(current_user, project, sample_params).execute
+
+      if sample.persisted?
+        sample
+      else
+        sample.errors.map do |error|
+          {
+            path: ['sample', error.attribute.to_s.camelize(:lower)],
+            message: error.message
+          }
+        end
+      end
+    end
+  end
+end

config/locales/en.yml

@@ -1675,6 +1675,14 @@ en:
         namespace_project_exists: Project with the same name or path already exists in the target namespace.
         project_in_namespace: Project in namespace
     samples:
+      batch_import:
+        project_puid_not_found: 'Project PUID %{project_puid} could not be found'
+        missing_field: 'Row with index %{index} is missing required fields'
+        missing_header: 'File is missing %{header_title} header for %{header_desc}'
+        missing_header_project: 'File is missing %{header_expected} header for project puid'
+        duplicate_column_names: The file has duplicate column header names. Please remove the columns with duplicate header names and retry uploading.
+        missing_data_row: The file does not have any data rows. Please add some rows of data to the file and retry uploading.
+        empty_file: The file cannot be empty.
       clone:
         empty_new_project_id: Destination project was not selected.
         empty_sample_ids: The sample ids are empty.

test/fixtures/files/valid_batch_sample_import.csv

@@ -0,0 +1,3 @@
+sample_name,project_puid,description
+my new sample,INXT_PRJ_AZCEZ6YMW3,my description
+my new sample 2,INXT_PRJ_AZCEZ6YMW3,my description 2