🎁 Add derivative_rodeo_splitter

Add a new PDF splitter option that wraps the DerivateRodeo's PdfSplitGenerator.

ref #220

lib/iiif_print.rb

@@ -19,6 +19,7 @@
 require "iiif_print/jobs/child_works_from_pdf_job"
 require "iiif_print/split_pdfs/base_splitter"
 require "iiif_print/split_pdfs/child_work_creation_from_pdf_service"
+require "iiif_print/split_pdfs/derivative_rodeo_splitter"
 
 module IiifPrint
   extend ActiveSupport::Autoload

lib/iiif_print/jobs/child_works_from_pdf_job.rb

@@ -15,17 +15,19 @@ def perform(candidate_for_parency, pdf_paths, user, admin_set_id, *)
         # We know that we have cases where parent_work is nil, this will definitely raise an
         # exception; which is fine because we were going to do it later anyway.
         @parent_work = if candidate_for_parency.work?
+                         pdf_file_set = nil
                          candidate_for_parency
                        else
                          # We likely have a file set
+                         pdf_file_set = candidate_for_parency
                          IiifPrint.parent_for(candidate_for_parency)
                        end
         @child_admin_set_id = admin_set_id
         child_model = @parent_work.iiif_print_config.pdf_split_child_model
 
-        # handle each input pdf
+        # handle each input pdf (when input is a file set, we will only have one).
         pdf_paths.each do |original_pdf_path|
-          split_pdf(original_pdf_path, user, child_model)
+          split_pdf(original_pdf_path, user, child_model, pdf_file_set)
         end
 
         # Link newly created child works to the parent
@@ -47,10 +49,8 @@ def perform(candidate_for_parency, pdf_paths, user, admin_set_id, *)
       private
 
       # rubocop:disable Metrics/ParameterLists
-      def split_pdf(original_pdf_path, user, child_model)
-        # TODO: This is the place to change out the existing service and instead use the derivative
-        # rodeo; we will likely need to look at method signatures to tighten this interface.
-        image_files = @parent_work.iiif_print_config.pdf_splitter_service.call(original_pdf_path)
+      def split_pdf(original_pdf_path, user, child_model, pdf_file_set)
+        image_files = @parent_work.iiif_print_config.pdf_splitter_service.call(original_pdf_path, file_set: pdf_file_set)
         return if image_files.blank?
 
         prepare_import_data(original_pdf_path, image_files, user)

lib/iiif_print/split_pdfs/derivative_rodeo_splitter.rb

@@ -0,0 +1,42 @@
+module IiifPrint
+  module SplitPdfs
+    class DerivativeRodeoSplitter
+      ##
+      # This class wraps the DerivativeRodeo::Generators::PdfSplitGenerator to find preprocessed
+      # images, or split a PDF if there are no preprocessed images.
+      #
+      # TODO: override output extension from default "tiff" in Derivative Rodeo
+      # TODO: define output_location_template & preprocessed_location_template
+
+      ##
+      # @param _path [String] unused here, kept for consistant splitter method signature
+      # @param file_set [FileSet] file set containing a PDF file to split
+      # @return [Array] paths to images split from each page of PDF file
+      def self.call(_path, file_set:)
+        new(file_set: file_set).split_files
+      end
+
+      def initialize(file_set:)
+        @path = IiifPrint::DerivativeRodeoService.derivative_rodeo_input_uri(file_set: file_set)
+      end
+
+      def split_files
+        DerivativeRodeo::Generators::PdfSplitGenerator.new(
+          input_uris: @path,
+          output_location_template: template,
+          preprocessed_location_template: location
+        ).generated_files
+      end
+
+      private
+
+      def template
+        'who knows'
+      end
+
+      def location
+        'who knows'
+      end
+    end
+  end
+end

spec/iiif_print/split_pdfs/derivative_rodeo_splitter_spec.rb

@@ -0,0 +1,33 @@
+# frozen_string_literal: true
+
+require 'spec_helper'
+
+RSpec.describe IiifPrint::SplitPdfs::DerivativeRodeoSplitter do
+  let(:path) { nil }
+  let(:work) { double(MyWork, aark_id: '12345') }
+  let(:file_set) { FileSet.new.tap { |fs| fs.save!(validate: false) } }
+
+  describe 'class' do
+    subject { described_class }
+
+    it { is_expected.to respond_to(:call) }
+  end
+
+  describe "instance" do
+    subject { described_class.new(file_set: file_set) }
+
+    before do
+      allow(file_set).to receive(:parent).and_return(work)
+      # TODO: This is a hack that leverages the internals of Hydra::Works; not excited about it but
+      # this part is only one piece of the over all integration.
+      allow(file_set).to receive(:original_file).and_return(double(original_filename: __FILE__))
+    end
+
+    it { is_expected.to respond_to :split_files }
+
+    it 'uses the rodeo to split' do
+      expect(DerivativeRodeo::Generators::PdfSplitGenerator).to receive(:new)
+      described_class.call(path, file_set: file_set)
+    end
+  end
+end