Merge pull request #1926 from samvera/iiif-print-demo-copy

Iiif print demo copy

Gemfile

@@ -83,7 +83,7 @@ group :development do
   gem 'spring-watcher-listen', '~> 2.0.0'
 end
 
-gem 'bulkrax', '~> 3.5.0'
+gem 'bulkrax', '~> 5.0'
 
 gem 'blacklight', '~> 6.7'
 gem 'blacklight_oai_provider', '~> 6.1', '>= 6.1.1'
@@ -92,7 +92,9 @@ gem 'hyrax', '~> 3.5.0'
 
 gem 'bolognese', '>= 1.9.10'
 gem 'hyrax-doi', git: 'https://github.com/samvera-labs/hyrax-doi.git', branch: 'hyrax_upgrade'
+gem 'iiif_print', git: 'https://github.com/scientist-softserv/iiif_print.git', branch: 'main'
 gem 'postrank-uri', '>= 1.0.24'
+gem 'redlock', '>= 0.1.2', '< 2.0' # lock redlock per https://github.com/samvera/hyrax/pull/5961
 gem 'rsolr', '~> 2.0'
 
 gem 'devise'

Gemfile.lock

@@ -9,6 +9,19 @@ GIT
       hyrax (>= 2.9, < 4.0)
       rails (~> 5.2.4, >= 5.2.4.3)
 
+GIT
+  remote: https://github.com/scientist-softserv/iiif_print.git
+  revision: 4f6b67276b57bf4648866408eb136905ca644e34
+  branch: main
+  specs:
+    iiif_print (0.0.1)
+      blacklight_iiif_search (~> 1.0)
+      dry-monads (~> 1.4.0)
+      hyrax (>= 2.5, < 4.0)
+      nokogiri (>= 1.13.2)
+      rails (~> 5.0)
+      rdf-vocab (~> 3.0)
+
 GIT
   remote: https://github.com/tawan/active-elastic-job.git
   revision: ec51c5d9dedc4a1b47f2db41f26d5fceb251e979
@@ -126,7 +139,7 @@ GEM
     babel-transpiler (0.7.0)
       babel-source (>= 4.0, < 6)
       execjs (~> 2.0)
-    bagit (0.4.4)
+    bagit (0.4.5)
       docopt (~> 0.5.0)
       validatable (~> 1.6)
     bcp47 (0.3.3)
@@ -155,6 +168,10 @@ GEM
       bootstrap-sass (~> 3.0)
       openseadragon (>= 0.2.0)
       rails
+    blacklight_iiif_search (1.0.0)
+      blacklight (~> 6.0)
+      iiif-presentation
+      rails (>= 4.2, < 6)
     blacklight_oai_provider (6.1.1)
       blacklight (~> 6.0)
       oai (~> 1.0)
@@ -200,13 +217,13 @@ GEM
       signet (~> 0.8)
       typhoeus
     builder (3.2.4)
-    bulkrax (3.5.1)
+    bulkrax (5.1.0)
       bagit (~> 0.4)
       coderay
       iso8601 (~> 0.9.0)
       kaminari
       language_list (~> 1.2, >= 1.2.1)
-      libxml-ruby (~> 3.1.0)
+      libxml-ruby (~> 3.2.4)
       loofah (>= 2.2.3)
       oai (>= 0.4, < 2.x)
       rack (>= 2.0.6)
@@ -548,6 +565,10 @@ GEM
       rainbow (>= 2.2.2, < 4.0)
       terminal-table (>= 1.5.1)
     ice_nine (0.11.2)
+    iiif-presentation (1.1.0)
+      activesupport (>= 3.2.18)
+      faraday (>= 0.9)
+      json
     iiif_manifest (1.3.0)
       activesupport (>= 4)
     is_it_working (1.1.0)
@@ -627,7 +648,7 @@ GEM
       rdf-vocab (~> 3.0)
     legato (0.7.0)
       multi_json
-    libxml-ruby (3.1.0)
+    libxml-ruby (3.2.4)
     link_header (0.0.8)
     linkeddata (3.1.6)
       equivalent-xml (~> 0.6)
@@ -1136,7 +1157,7 @@ DEPENDENCIES
   blacklight_oai_provider (~> 6.1, >= 6.1.1)
   bolognese (>= 1.9.10)
   bootstrap-datepicker-rails
-  bulkrax (~> 3.5.0)
+  bulkrax (~> 5.0)
   byebug
   capybara
   capybara-screenshot (~> 1.0)
@@ -1159,6 +1180,7 @@ DEPENDENCIES
   hyrax-doi!
   i18n-debug
   i18n-tasks
+  iiif_print!
   is_it_working
   jbuilder (~> 2.5)
   jquery-rails
@@ -1176,6 +1198,7 @@ DEPENDENCIES
   rails (~> 5.2.5)
   rails-controller-testing
   rdf (~> 3.1.15)
+  redlock (>= 0.1.2, < 2.0)
   riiif (~> 1.1)
   rolify
   rsolr (~> 2.0)

app/assets/javascripts/application.js

@@ -43,6 +43,8 @@
 //= require bulkrax/application
 
 //= require hyrax
+//= require iiif_print
+
 //= require jquery.flot.pie
 //= require flot_graph
 //= require statistics_tab_manager

app/assets/stylesheets/iiif_print.scss

@@ -0,0 +1 @@
+@import 'iiif_print/iiif_print';

app/controllers/catalog_controller.rb

@@ -16,7 +16,22 @@ def self.modified_field
     'system_modified_dtsi'
   end
 
+  # CatalogController-scope behavior and configuration for BlacklightIiifSearch
+  include BlacklightIiifSearch::Controller
+
   configure_blacklight do |config|
+    # IiifPrint index fields
+    config.add_index_field 'all_text_tsimv', highlight: true, helper_method: :render_ocr_snippets
+
+    # configuration for Blacklight IIIF Content Search
+    config.iiif_search = {
+      full_text_field: 'all_text_tsimv',
+      object_relation_field: 'is_page_of_ssim',
+      supported_params: %w[q page],
+      autocomplete_handler: 'iiif_suggest',
+      suggester_name: 'iiifSuggester'
+    }
+
     config.view.gallery.partials = %i[index_header index]
     config.view.masonry.partials = [:index]
     config.view.slideshow.partials = [:index]
@@ -30,7 +45,7 @@ def self.modified_field
     config.advanced_search[:query_parser] ||= 'dismax'
     config.advanced_search[:form_solr_parameters] ||= {}
 
-    config.search_builder_class = Hyrax::CatalogSearchBuilder
+    config.search_builder_class = IiifPrint::CatalogSearchBuilder
 
     # Show gallery view
     config.view.gallery.partials = %i[index_header index]

app/models/ability.rb

@@ -101,4 +101,12 @@ def all_user_and_group_roles
   def featured_collection_abilities
     can %i[create destroy update], FeaturedCollection if admin?
   end
+
+  def can_import_works?
+    can_create_any_work?
+  end
+
+  def can_export_works?
+    can_create_any_work?
+  end
 end

app/models/generic_work.rb

@@ -3,6 +3,9 @@
 class GenericWork < ActiveFedora::Base
   include ::Hyrax::WorkBehavior
   include ::Hyrax::BasicMetadata
+  include IiifPrint.model_configuration(
+    pdf_split_child_model: self
+  )
 
   validates :title, presence: { message: 'Your work must have a title.' }
 

app/models/iiif_search_builder.rb

@@ -0,0 +1,17 @@
+# frozen_string_literal: true
+
+# SearchBuilder for full-text searches with highlighting and snippets
+class IiifSearchBuilder < Blacklight::SearchBuilder
+  include Blacklight::Solr::SearchBuilderBehavior
+
+  self.default_processor_chain += [:ocr_search_params]
+
+  # set params for ocr field searching
+  def ocr_search_params(solr_parameters = {})
+    solr_parameters[:facet] = false
+    solr_parameters[:hl] = true
+    solr_parameters[:'hl.fl'] = blacklight_config.iiif_search[:full_text_field]
+    solr_parameters[:'hl.fragsize'] = 100
+    solr_parameters[:'hl.snippets'] = 10
+  end
+end

app/models/image.rb

@@ -4,6 +4,9 @@
 #  `rails generate hyrax:work Image`
 class Image < ActiveFedora::Base
   include ::Hyrax::WorkBehavior
+  include IiifPrint.model_configuration(
+    pdf_split_child_model: self
+  )
 
   property :extent, predicate: ::RDF::Vocab::DC.extent, multiple: true do |index|
     index.as :stored_searchable

config/environments/development.rb

@@ -40,7 +40,7 @@
   # Debug mode disables concatenation and preprocessing of assets.
   # This option may cause significant delays in view rendering with a large
   # number of complex assets.
-  config.assets.debug = true
+  # config.assets.debug = true
 
   # Suppress logger output for asset requests.
   config.assets.quiet = true

config/initializers/bulkrax.rb

@@ -61,4 +61,6 @@
     # Properties that should not be used in imports/exports. They are reserved for use by Hyrax.
     # config.reserved_properties += ['my_field']
   end
+
+  Bulkrax::CreateRelationshipsJob.update_child_records_works_file_sets = true
 end

config/initializers/iiif_print.rb

@@ -0,0 +1,22 @@
+IiifPrint.config do |config|
+  # NOTE: WorkTypes and models are used synonymously here.
+  # Add models to be excluded from search so the user
+  # would not see them in the search results.
+  # by default, use the human readable versions like:
+  # @example
+  #   # config.excluded_model_name_solr_field_values = ['Generic Work', 'Image']
+  #
+  # config.excluded_model_name_solr_field_values = []
+
+  # Add configurable solr field key for searching,
+  # default key is: 'human_readable_type_sim'
+  # if another key is used, make sure to adjust the
+  # config.excluded_model_name_solr_field_values to match
+  # @example
+  #   config.excluded_model_name_solr_field_key = 'some_solr_field_key'
+
+  # Configure how the manifest sorts the canvases, by default it sorts by :title,
+  # but a different model property may be desired such as :date_published
+  # @example
+  #   config.sort_iiif_manifest_canvases_by = :date_published
+end

config/routes.rb

@@ -3,6 +3,8 @@
 require 'sidekiq/web'
 
 Rails.application.routes.draw do
+
+  concern :iiif_search, BlacklightIiifSearch::Routes.new
   concern :oai_provider, BlacklightOaiProvider::Routes.new
 
   mount Riiif::Engine => 'images', as: :riiif if Hyrax.config.iiif_image_server?
@@ -59,6 +61,7 @@
 
   resources :solr_documents, only: [:show], path: '/catalog', controller: 'catalog' do
     concerns :exportable
+    concerns :iiif_search
   end
 
   resources :bookmarks do

db/migrate/20221205210847_create_iiif_print_derivative_attachments.iiif_print.rb

@@ -0,0 +1,13 @@
+# This migration comes from iiif_print (originally 20181214181358)
+class CreateIiifPrintDerivativeAttachments < ActiveRecord::Migration[5.0]
+  def change
+    create_table :iiif_print_derivative_attachments do |t|
+      t.string :fileset_id
+      t.string :path
+      t.string :destination_name
+
+      t.timestamps
+    end
+    add_index :iiif_print_derivative_attachments, :fileset_id
+  end
+end

db/migrate/20221205210848_create_iiif_print_ingest_file_relations.iiif_print.rb

@@ -0,0 +1,12 @@
+# This migration comes from iiif_print (originally 20190107165909)
+class CreateIiifPrintIngestFileRelations < ActiveRecord::Migration[5.0]
+  def change
+    create_table :iiif_print_ingest_file_relations do |t|
+      t.string :file_path
+      t.string :derivative_path
+
+      t.timestamps
+    end
+    add_index :iiif_print_ingest_file_relations, :file_path
+  end
+end

db/migrate/20230131202855_create_iiif_print_pending_relationships.iiif_print.rb

@@ -0,0 +1,12 @@
+# This migration comes from iiif_print (originally 20230109000000)
+class CreateIiifPrintPendingRelationships < ActiveRecord::Migration[5.1]
+  def change
+    create_table :iiif_print_pending_relationships do |t|
+      t.string :child_title, null: false
+      t.string :parent_id, null: false
+      t.string :child_order, null: false
+      t.timestamps
+    end
+    add_index :iiif_print_pending_relationships, :parent_id
+  end
+end

db/schema.rb

@@ -10,7 +10,7 @@
 #
 # It's strongly recommended that you check this file into your version control system.
 
-ActiveRecord::Schema.define(version: 2022_11_22_001715) do
+ActiveRecord::Schema.define(version: 2023_02_08_001228) do
 
   # These are extensions that must be enabled in order to support this database
   enable_extension "plpgsql"
@@ -351,6 +351,32 @@
     t.string "humanized_name"
   end
 
+  create_table "iiif_print_derivative_attachments", id: :serial, force: :cascade do |t|
+    t.string "fileset_id"
+    t.string "path"
+    t.string "destination_name"
+    t.datetime "created_at", null: false
+    t.datetime "updated_at", null: false
+    t.index ["fileset_id"], name: "index_iiif_print_derivative_attachments_on_fileset_id"
+  end
+
+  create_table "iiif_print_ingest_file_relations", id: :serial, force: :cascade do |t|
+    t.string "file_path"
+    t.string "derivative_path"
+    t.datetime "created_at", null: false
+    t.datetime "updated_at", null: false
+    t.index ["file_path"], name: "index_iiif_print_ingest_file_relations_on_file_path"
+  end
+
+  create_table "iiif_print_pending_relationships", force: :cascade do |t|
+    t.string "child_title", null: false
+    t.string "parent_id", null: false
+    t.string "child_order", null: false
+    t.datetime "created_at", null: false
+    t.datetime "updated_at", null: false
+    t.index ["parent_id"], name: "index_iiif_print_pending_relationships_on_parent_id"
+  end
+
   create_table "job_io_wrappers", id: :serial, force: :cascade do |t|
     t.integer "user_id"
     t.integer "uploaded_file_id"

solr/conf/schema.xml

@@ -171,7 +171,24 @@
         <filter class="solr.RemoveDuplicatesTokenFilterFactory"/>
       </analyzer>
     </fieldType>
-  </types>
+
+    <!-- BEGIN Blacklight IIIF Search autocomplete config -->
+    <!-- TODO needs configuration work to enable autocomplete -->
+    <!-- <fieldType name="textSuggestTokenizer" class="solr.TextField" positionIncrementGap="100"> -->
+    <!-- <analyzer> -->
+    <!-- <tokenizer class="solr.WhitespaceTokenizerFactory"/> -->
+    <!-- <filter class="solr.StopFilterFactory" ignoreCase="true" words="stopwords_en.txt"/> -->
+    <!-- <filter class="solr.WordDelimiterGraphFilterFactory"/> -->
+    <!-- <filter class="solr.LowerCaseFilterFactory"/> -->
+    <!-- <filter class="solr.HyphenatedWordsFilterFactory"/> -->
+    <!-- <filter class="solr.RemoveDuplicatesTokenFilterFactory"/> -->
+        <!-- uncomment below to enable multi-word matches -->
+        <!-- <filter class="solr.ShingleFilterFactory" outputUnigrams="true" outputUnigramsIfNoShingles="true" maxShingleSize="3" /> -->
+        <!-- </analyzer> -->
+        <!-- </fieldType> -->
+    <!-- END Blacklight IIIF Search autocomplete config -->
+
+</types>
 
   <fields>
     <!-- If you remove this field, you must _also_ disable the update log in solrconfig.xml
@@ -323,7 +340,11 @@
     <!-- you must define copyField source and dest fields explicity or schemaBrowser doesn't work -->
     <field name="all_text_timv" type="text" stored="false" indexed="true" multiValued="true" termVectors="true" termPositions="true" termOffsets="true"/>
 
-  </fields>
+    <!-- BEGIN Blacklight IIIF Search autocomplete config -->
+    <!-- <field name="iiif_suggest" type="textSuggest" indexed="true" stored="true" multiValued="true" /> -->
+   <!-- END Blacklight IIIF Search autocomplete config -->
+
+</fields>
 
  <!-- Field to use to determine and enforce document uniqueness.
       Unless this field is marked with required="false", it will be a required field
@@ -359,4 +380,8 @@
  </similarity>
  -->
 
+  <!-- BEGIN Blacklight IIIF Search autocomplete config -->
+  <!-- TODO this needs work <copyField source="all_text_tsimv" dest="iiif_suggest"/> -->
+  <!-- END Blacklight IIIF Search autocomplete config -->
+
 </schema>