iiif_print 1.0.0 → 2.0.0

data/lib/iiif_print/jobs/child_works_from_pdf_job.rb

@@ -1,107 +0,0 @@
-module IiifPrint
-  module Jobs
-    class ChildWorksFromPdfJob < IiifPrint::Jobs::ApplicationJob
-      # Break a pdf into individual pages
-      # @param parent_work
-      # @param pdf_paths: [<Array => String>] paths to pdfs
-      # @param user: [User]
-      # @param admin_set_id: [<String>]
-      # @param prior_pdfs: [<Integer>] count of pdfs already on parent work
-      def perform(parent_work, pdf_paths, user, admin_set_id, prior_pdfs)
-        @parent_work = parent_work
-        @child_admin_set_id = admin_set_id
-        child_model = @parent_work.iiif_print_config.pdf_split_child_model
-
-        # handle each input pdf
-        pdf_paths.each_with_index do |path, pdf_idx|
-          split_pdf(path, pdf_idx, user, prior_pdfs, child_model)
-        end
-
-        # Link newly created child works to the parent
-        # @param user: [User] user
-        # @param parent_id: [<String>] parent work id
-        # @param parent_model: [<String>] parent model
-        # @param child_model: [<String>] child model
-        IiifPrint::Jobs::CreateRelationshipsJob.set(wait: 10.minutes).perform_later(
-          user: user,
-          parent_id: @parent_work.id,
-          parent_model: @parent_work.class.to_s,
-          child_model: child_model.to_s
-        )
-
-        # TODO: clean up image_files and pdf_paths
-      end
-
-      private
-
-      def split_pdf(path, pdf_idx, user, prior_pdfs_count, child_model)
-        image_files = @parent_work.iiif_print_config.pdf_splitter_service.new(path).to_a
-        return if image_files.blank?
-
-        pdf_sequence = pdf_idx + prior_pdfs_count
-        prepare_import_data(pdf_sequence, image_files, user)
-
-        # submit the job to create all the child works for one PDF
-        # @param [User] user
-        # @param [Hash<String => String>] titles
-        # @param [Hash<String => String>] resource_types (optional)
-        # @param [Array<String>] uploaded_files Hyrax::UploadedFile IDs
-        # @param [Hash] attributes attributes to apply to all works, including :model
-        # @param [Hyrax::BatchCreateOperation] operation
-        operation = Hyrax::BatchCreateOperation.create!(
-          user: user,
-          operation_type: "PDF Batch Create"
-        )
-        BatchCreateJob.perform_later(user,
-                                     @child_work_titles,
-                                     {},
-                                     @uploaded_files,
-                                     attributes.merge!(model: child_model.to_s).with_indifferent_access,
-                                     operation)
-      end
-
-      def prepare_import_data(pdf_sequence, image_files, user)
-        @uploaded_files = []
-        @child_work_titles = {}
-        image_files.each_with_index do |image_path, idx|
-          file_id = create_uploaded_file(user, image_path).to_s
-          file_title = set_title(@parent_work.title.first, pdf_sequence, idx)
-          @uploaded_files << file_id
-          @child_work_titles[file_id] = file_title
-          # save child work info to create the member relationships
-          PendingRelationship.create!(child_title: file_title,
-                                      parent_id: @parent_work.id,
-                                      child_order: sort_order(pdf_sequence, idx))
-        end
-      end
-
-      def sort_order(pdf_sequence, idx)
-        "#{pdf_sequence} #{idx}"
-      end
-
-      def create_uploaded_file(user, path)
-        uf = Hyrax::UploadedFile.new
-        uf.user_id = user.id
-        uf.file = CarrierWave::SanitizedFile.new(path)
-        uf.save!
-        uf.id
-      end
-
-      def set_title(title, pdf_sequence, idx)
-        pdf_index = "Pdf Nbr #{pdf_sequence + 1}"
-        page_number = "Page #{idx + 1}"
-        "#{title}: #{pdf_index}, #{page_number}"
-      end
-
-      # TODO: what attributes do we need to fill in from the parent work? What about AllinsonFlex?
-      def attributes
-        {
-          admin_set_id: @child_admin_set_id.to_s,
-          creator: @parent_work.creator.to_a,
-          rights_statement: @parent_work.rights_statement.to_a,
-          visibility: @parent_work.visibility.to_s
-        }
-      end
-    end
-  end
-end

data/lib/iiif_print/jobs/create_relationships_job.rb

@@ -1,78 +0,0 @@
-module IiifPrint
-  module Jobs
-    # Break a pdf into individual pages
-    class CreateRelationshipsJob < IiifPrint::Jobs::ApplicationJob
-      # Link newly created child works to the parent
-      # @param user: [User] user
-      # @param parent_id: [<String>] parent work id
-      # @param parent_model: [<String>] parent model
-      # @param child_model: [<String>] child model
-      def perform(user:, parent_id:, parent_model:, child_model:)
-        if completed_child_data_for(parent_id, child_model)
-          # add the members
-          parent_work = parent_model.constantize.find(parent_id)
-          create_relationships(user: user, parent: parent_work, ordered_children: @child_works)
-          @pending_children.each(&:destroy)
-        else
-          # reschedule the job and end this one normally
-          #
-          # TODO: Depending on how things shake out, we could be infinitely rescheduling this job.
-          # Consider a time to live parameter.
-          reschedule(user: user, parent_id: parent_id, parent_model: parent_model, child_model: child_model)
-        end
-      end
-
-      private
-
-      # load @child_works, and return true or false
-      def completed_child_data_for(parent_id, child_model)
-        @child_works = []
-        found_all_children = true
-
-        # find and sequence all pending children
-        @pending_children = IiifPrint::PendingRelationship.where(parent_id: parent_id).order('child_order asc')
-
-        # find child works (skip out if any haven't yet been created)
-        @pending_children.each do |child|
-          # find by title... if any aren't found, the child works are not yet ready
-          found_children = find_children_by_title_for(child.child_title, child_model)
-          found_all_children = false if found_children.empty?
-          break unless found_all_children == true
-          @child_works += found_children
-        end
-        # return boolean
-        found_all_children
-      end
-
-      def find_children_by_title_for(title, model)
-        # We should only find one, but there is no guarantee of that and `:where` returns an array.
-        model.constantize.where(title: title)
-      end
-
-      def reschedule(user:, parent_id:, parent_model:, child_model:)
-        CreateRelationshipsJob.set(wait: 10.minutes).perform_later(
-          user: user,
-          parent_id: parent_id,
-          parent_model: parent_model,
-          child_model: child_model
-        )
-      end
-
-      def create_relationships(user:, parent:, ordered_children:)
-        records_hash = {}
-        ordered_children.map(&:id).each_with_index do |child_id, i|
-          records_hash[i.to_s] = { id: child_id }
-        end
-        attrs = { work_members_attributes: records_hash }
-        parent.try(:reindex_extent=, Hyrax::Adapters::NestingIndexAdapter::LIMITED_REINDEX)
-        env = Hyrax::Actors::Environment.new(parent, Ability.new(user), attrs)
-
-        Hyrax::CurationConcern.actor.update(env)
-        # need to reindex all file_sets to make all ancestors are indexed
-        ordered_children.each do |child_work|
-          child_work.file_sets.each(&:update_index) if child_work.respond_to?(:file_sets)
-        end
-      end
-    end
-  end
-end

data/lib/iiif_print/split_pdfs/pages_into_images_service.rb

@@ -1,130 +0,0 @@
-require 'open3'
-require 'securerandom'
-require 'tmpdir'
-require 'iiif_print/split_pdfs/pdf_image_extraction_service'
-
-module IiifPrint
-  module SplitPdfs
-    class PagesIntoImagesService
-      include Enumerable
-
-      def initialize(path)
-        @baseid = SecureRandom.uuid
-        @pdfpath = path
-        @info = nil
-        @entries = nil
-        @tmpdir = nil
-        @size = nil
-        @pagecount = nil
-        @pdftext = nil
-        @compression = 'lzw'
-      end
-
-      # return
-      def pdfinfo
-        @info = IiifPrint::SplitPdfs::PdfImageExtractionService.new(@pdfpath) if @info.nil?
-        @info
-      end
-
-      # TODO: put this test somewhere to prevent invalid pdfs from crashing the image service.
-      def invalid_pdf?
-        return true if pdfinfo.color.include?(nil) || pdfinfo.width.nil? || pdfinfo.height.nil? || pdfinfo.entries.length.zero?
-        false
-      end
-
-      def tmpdir
-        @tmpdir = Dir.mktmpdir if @tmpdir.nil?
-        @tmpdir
-      end
-
-      def colordevice(channels, bpc)
-        bits = bpc * channels
-        # will be either 8bpc/16bpd color TIFF,
-        #   with any CMYK source transformed to 8bpc RBG
-        bits = 24 unless [24, 48].include? bits
-        "tiff#{bits}nc"
-      end
-
-      def gsdevice
-        color, channels, bpc = pdfinfo.color
-        device = nil
-        # CCITT Group 4 Black and White, if applicable:
-        if color == 'gray' && bpc == 1
-          device = 'tiffg4'
-          @compression = 'g4'
-        end
-        # 8 Bit Grayscale, if applicable:
-        device = 'tiffgray' if color == 'gray' && bpc > 1
-        # otherwise color:
-        device = colordevice(channels, bpc) if device.nil?
-        device
-      end
-
-      # TODO: this method came from newspaper gem but appears to be unused. Is it needed anywhere?
-      # def gstext
-      #   cmd = "gs -q -dNOPAUSE -dBATCH -sDEVICE=txtwrite " \
-      #         "-sOutputFile=- -f #{@pdfpath}"
-      #   Open3.popen3(cmd) do |_stdin, stdout, _stderr, _wait_thr|
-      #     @pdftext = stdout.read
-      #   end
-      #   @pdftext
-      # end
-
-      def pagecount
-        cmd = "pdfinfo #{@pdfpath}"
-        Open3.popen3(cmd) do |_stdin, stdout, _stderr, _wait_thr|
-          output = stdout.read.split("\n")
-          # rubocop:disable Performance/Detect
-          pages_e = output.select { |e| e.start_with?('Pages:') }[0]
-          # rubocop:enable Performance/Detect
-          @pagecount = pages_e.split[-1].to_i
-        end
-        @pagecount
-      end
-
-      def looks_scanned
-        max_image_px = pdfinfo.width * pdfinfo.height
-        single_image_per_page = pdfinfo.entries.length == pagecount
-        # single 10mp+ image per page?
-        single_image_per_page && max_image_px > 1024 * 1024 * 10
-      end
-
-      def ppi
-        unless looks_scanned
-          # 400 dpi for something that does not look like scanned media:
-          return 400
-        end
-        # For scanned media, defer to detected image PPI:
-        pdfinfo.ppi
-      end
-
-      # ghostscript convert all pages to TIFF
-      def gsconvert
-        output_base = File.join(tmpdir, "#{@baseid}-page%d.tiff")
-        cmd = "gs -dNOPAUSE -dBATCH -sDEVICE=#{gsdevice} " \
-              "-dTextAlphaBits=4 -sCompression=#{@compression} " \
-              "-sOutputFile=#{output_base} -r#{ppi} -f #{@pdfpath}"
-        Open3.popen3(cmd) do |_stdin, stdout, _stderr, _wait_thr|
-          output = stdout.read.split("\n")
-          # rubocop:disable Performance/Count
-          @size = output.select { |e| e.start_with?('Page ') }.length
-          # rubocop:enable Performance/Count
-        end
-        # Return an array of expected filenames
-        (1..@size).map { |n| File.join(tmpdir, "#{@baseid}-page#{n}.tiff") }
-      end
-
-      # entries for each page
-      def entries
-        @entries = gsconvert if @entries.nil?
-        @entries
-      end
-
-      def each
-        entries.each do |e|
-          yield(e)
-        end
-      end
-    end
-  end
-end

data/spec/.keep.txt

@@ -1 +0,0 @@
-spec dir for RSpec

data/spec/factories/ability.rb

@@ -1,6 +0,0 @@
-FactoryBot.define do
-  factory :ability do
-    user
-    initialize_with { new(user) }
-  end
-end

data/spec/factories/newspaper_issue.rb

@@ -1,7 +0,0 @@
-# will infer, create a NewspaperIssue object
-FactoryBot.define do
-  factory :newspaper_issue do
-    title { ['Here and There'] }
-    depositor { User.batch_user.user_key }
-  end
-end

data/spec/factories/newspaper_page.rb

@@ -1,7 +0,0 @@
-# will infer, create a NewspaperPage object
-FactoryBot.define do
-  factory :newspaper_page do
-    title { ['Here and There'] }
-    depositor { User.batch_user.user_key }
-  end
-end

data/spec/factories/newspaper_page_solr_document.rb

@@ -1,12 +0,0 @@
-FactoryBot.define do
-  factory :newspaper_page_solr_document, class: SolrDocument do
-    initialize_with do
-      new(id: '123456',
-          title_tesim: ['Page 1'],
-          has_model_ssim: ['NewspaperPage'],
-          issue_id_ssi: 'abc123',
-          file_set_ids_ssim: ['7891011'],
-          thumbnail_path_ss: '/downloads/123456?file=thumbnail')
-    end
-  end
-end

data/spec/factories/newspaper_title.rb

@@ -1,8 +0,0 @@
-# will infer, create a NewspaperTitle object
-FactoryBot.define do
-  factory :newspaper_title do
-    title { ['ACME Press'] }
-    lccn { 'sn2036999999' }
-    depositor { User.batch_user.user_key }
-  end
-end

data/spec/factories/uploaded_pdf_file.rb

@@ -1,9 +0,0 @@
-FactoryBot.define do
-  factory :uploaded_pdf_file, class: Hyrax::UploadedFile do
-    initialize_with do
-      base = File.join(IiifPrint::GEM_PATH, 'spec', 'fixtures', 'files')
-      pdf_path = File.join(base, 'minimal-2-page.pdf')
-      new(file: File.open(pdf_path), user: create(:user))
-    end
-  end
-end

data/spec/factories/uploaded_txt_file.rb

@@ -1,9 +0,0 @@
-FactoryBot.define do
-  factory :uploaded_txt_file, class: Hyrax::UploadedFile do
-    initialize_with do
-      base = File.join(IiifPrint::GEM_PATH, 'spec', 'fixtures', 'files')
-      file_path = File.join(base, 'ndnp-sample1-txt.txt')
-      new(file: File.open(file_path), user: create(:user))
-    end
-  end
-end

data/spec/factories/user.rb

@@ -1,13 +0,0 @@
-FactoryBot.define do
-  factory :user do
-    id { "skroob" }
-    email { "spaceballs@example.com" }
-    password { "password_is_12345" }
-    initialize_with do
-      User.find_or_create_by(id: id) do |user|
-        user.email = email
-        user.password = password
-      end
-    end
-  end
-end

data/spec/fixtures/files/README.md

@@ -1,7 +0,0 @@
-## Sample / fixture file manifest
-
-* sample-4page-issue.pdf
-  - original: chicopee-weekly-journal_1856-05-24-p1.pdf
-* page1.tiff
-  - first page from sample-4page-issue.pdf, converted to CCITT Group 4
-    encoded (lossless compressed) TIFF via Ghostscript.