I350 Derivatives & Update IiifPrint (#2210)

* Update IiifPrint * File Set Specs Update specs based on derivative & file migration processing in Hyrax. * Update IiifPrint to released gem * Bring in most recent Hyrax updates * Remove solr_document_behavior_decorator Hyrax now redevines hydra_model so we don't have to.
samvera · Jun 7, 2024 · 53cf2e8 · 53cf2e8
1 parent e046dd7
commit 53cf2e8
Show file tree

Hide file tree

Showing 19 changed files with 137 additions and 220 deletions.
diff --git a/Dockerfile b/Dockerfile
@@ -46,6 +46,11 @@ RUN wget https://github.com/ImageMagick/ImageMagick/archive/refs/tags/7.1.0-57.t
     && rm -rf ImageMagick* \
     && rm -rf /var/cache/apk/*
 
+# Install "best" training data for Tesseract
+RUN echo "📚 Installing Tesseract Best (training data)!" && \
+    cd /usr/share/tessdata/ && \
+    wget https://github.com/tesseract-ocr/tessdata_best/blob/main/eng.traineddata?raw=true -O eng_best.traineddata
+
 ARG VIPS_VERSION=8.11.3
 
 RUN set -x -o pipefail \

diff --git a/Gemfile b/Gemfile
@@ -56,7 +56,7 @@ gem 'hyrax-doi', github: 'samvera-labs/hyrax-doi', branch: 'rails_hyrax_upgrade'
 gem 'hyrax-iiif_av', github: 'samvera-labs/hyrax-iiif_av', branch: 'rails_hyrax_upgrade'
 gem 'i18n-debug', require: false, group: %i[development test]
 gem 'i18n-tasks', group: %i[development test]
-gem 'iiif_print', github: 'scientist-softserv/iiif_print', branch: 'main'
+gem 'iiif_print', '~> 2.0'
 gem 'jbuilder', '~> 2.5'
 gem 'jquery-rails' # Use jquery as the JavaScript library
 gem 'openssl', '>= 3.2.0'

diff --git a/Gemfile.lock b/Gemfile.lock
@@ -147,7 +147,7 @@ GIT
 
 GIT
   remote: https://github.com/samvera/hyrax.git
-  revision: 151fb0d38339e12eabdf99ce9d0b1a0f07685fd5
+  revision: 74f29943d88749a57077818d271b29729a6d7183
   branch: double_combo
   specs:
     hyrax (5.0.1)
@@ -218,18 +218,6 @@ GIT
       public_suffix (>= 2)
       rack (>= 1.3.6)
 
-GIT
-  remote: https://github.com/scientist-softserv/iiif_print.git
-  revision: 122e86634f17c8f40d9ae4d59d2d7deca00f1694
-  branch: main
-  specs:
-    iiif_print (1.0.0)
-      blacklight_iiif_search (>= 1.0, < 3.0)
-      derivative-rodeo (~> 0.5)
-      hyrax (>= 2.5, < 6)
-      nokogiri (>= 1.13.2)
-      rdf-vocab (~> 3.0)
-
 GIT
   remote: https://github.com/scientist-softserv/willow_sword.git
   revision: 38a0906647fae2020e8b0b08e296f85c457fcb34
@@ -695,7 +683,8 @@ GEM
     hiredis (0.6.3)
     htmlentities (4.3.4)
     http_logger (0.7.0)
-    httparty (0.21.0)
+    httparty (0.22.0)
+      csv
       mini_mime (>= 1.0.0)
       multi_xml (>= 0.5.2)
     httpclient (2.8.3)
@@ -767,6 +756,12 @@ GEM
       json
     iiif_manifest (1.3.1)
       activesupport (>= 4)
+    iiif_print (2.0.0)
+      blacklight_iiif_search (>= 1.0, < 3.0)
+      derivative-rodeo (~> 0.5)
+      hyrax (>= 2.5, < 6)
+      nokogiri (>= 1.13.2)
+      rdf-vocab (~> 3.0)
     iso-639 (0.3.6)
     iso8601 (0.9.1)
     jaro_winkler (1.5.6)
@@ -1504,7 +1499,7 @@ DEPENDENCIES
   hyrax-iiif_av!
   i18n-debug
   i18n-tasks
-  iiif_print!
+  iiif_print (~> 2.0)
   jbuilder (~> 2.5)
   jquery-rails
   json-canonicalization (= 0.3.1)

diff --git a/app/models/ability.rb b/app/models/ability.rb
@@ -110,4 +110,23 @@ def can_import_works?
   def can_export_works?
     can_create_any_work?
   end
+
+  ##
+  # @api public
+  #
+  # Overrides hydra-head, (and restores the method from blacklight-access-controls)
+  def download_permissions
+    can :download, [::String, ::Valkyrie::ID] do |id|
+      test_download(id.to_s)
+    end
+
+    can :download, ::SolrDocument do |obj|
+      if obj.pdf? && !obj.show_pdf_download_button
+        false
+      else
+        cache.put(obj.id, obj)
+        test_download(obj.id)
+      end
+    end
+  end
 end
diff --git a/app/models/concerns/hyrax/solr_document_behavior_decorator.rb b/app/models/concerns/hyrax/solr_document_behavior_decorator.rb
diff --git a/app/services/iiif_print/tenant_config.rb b/app/services/iiif_print/tenant_config.rb
@@ -100,6 +100,11 @@ module PdfSplitter
       mattr_accessor :iiif_print_splitter
       self.iiif_print_splitter = ::IiifPrint::SplitPdfs::PagesToJpgsSplitter
 
+      ##
+      def self.never_split_pdfs?
+        !TenantConfig.use_iiif_print?
+      end
+
       ##
       # @api public
       def self.call(*args)
@@ -147,6 +152,12 @@ def service
     # In Hyrax::WorkShowPresenter we're only looking at the underlying file_sets.  But IiifPrint
     # needs to look at multiple places.
     module WorkShowPresenterDecorator
+      ##
+      # @return [Boolean] Identifies whether IiifPrint PDF splitting is active for this work's tenant
+      def split_pdfs?
+        TenantConfig.use_iiif_print?
+      end
+
       ##
       # @return [Array<Symbol>] predicate methods (e.g. ending in "?") that reflect the types
       #         of files we want to consider for showing in the IIIF Viewer.

diff --git a/app/views/hyrax/file_sets/_actions.html.erb b/app/views/hyrax/file_sets/_actions.html.erb
diff --git a/config/initializers/iiif_print.rb b/config/initializers/iiif_print.rb
@@ -22,4 +22,5 @@
   #   config.sort_iiif_manifest_canvases_by = :date_published
   config.default_iiif_manifest_version = 3
   config.persistence_adapter = IiifPrint::PersistenceLayer::ValkyrieAdapter
+  config.additional_tesseract_options = "-l eng_best"
 end
diff --git a/...igrate/20240214005253_add_model_details_to_iiif_print_pending_relationships.iiif_print.rb b/...igrate/20240214005253_add_model_details_to_iiif_print_pending_relationships.iiif_print.rb
@@ -1,8 +1,8 @@
 # This migration comes from iiif_print (originally 20231110163052)
 class AddModelDetailsToIiifPrintPendingRelationships < ActiveRecord::Migration[5.2]
   def change
-    add_column :iiif_print_pending_relationships, :parent_model, :string
-    add_column :iiif_print_pending_relationships, :child_model, :string
-    add_column :iiif_print_pending_relationships, :file_id, :string
+    add_column :iiif_print_pending_relationships, :parent_model, :string unless column_exists?(:iiif_print_pending_relationships, :parent_model)
+    add_column :iiif_print_pending_relationships, :child_model, :string  unless column_exists?(:iiif_print_pending_relationships, :child_model)
+    add_column :iiif_print_pending_relationships, :file_id, :string  unless column_exists?(:iiif_print_pending_relationships, :file_id)
   end
 end
diff --git a/docker-compose.bundle.yml b/docker-compose.bundle.yml
@@ -1,6 +1,5 @@
 # Copy file to docker-compose.override.yml to override docker-compose.yml
 # Only use for local development
-version: '3.8'
 services:
   # Uncomment to allow for the use of a ruby debugger (byebug, pry, etc) in Docker.
   # See http://playbook-staging.notch8.com/en/devops/docker_debugger for more info.

diff --git a/docker-compose.override-nodory.yml b/docker-compose.override-nodory.yml
@@ -1,5 +1,3 @@
-version: '3.8'
-
 services:
   solr:
     ports:

diff --git a/docker-compose.production.yml b/docker-compose.production.yml
@@ -1,5 +1,3 @@
-version: '3.8'
-
 x-app: &app
   image: ghcr.io/samvera/hyku:${TAG:-latest}
   env_file:

diff --git a/docker-compose.yml b/docker-compose.yml
@@ -1,4 +1,3 @@
-version: '3.8'
 x-app: &app
   build:
     context: .

diff --git a/spec/fixtures/derivatives/00-xml.xml b/spec/fixtures/derivatives/00-xml.xml
@@ -0,0 +1,23 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<alto xmlns="http://www.loc.gov/standards/alto/ns-v2#">
+  <Description>
+    <MeasurementUnit>pixel</MeasurementUnit>
+  </Description>
+  <Layout>
+    <Page ID="ID1" PHYSICAL_IMG_NR="1" HEIGHT="3685" WIDTH="2457">
+      <PrintSpace HEIGHT="3685" WIDTH="2457" HPOS="0" VPOS="0">
+        <TextBlock ID="ID1a" HEIGHT="3685" WIDTH="2457" HPOS="0" VPOS="0">
+          <TextLine HEIGHT="3685" WIDTH="2457" HPOS="0" VPOS="0">
+            <String CONTENT=" &#10;     &#10;    &#10;   &#10;  &#10; &#10;&#10;" WIDTH="1616" HEIGHT="3128" HPOS="466" VPOS="276"></String>
+            <String CONTENT=" &#10;     &#10;    &#10;   &#10;  &#10; &#10;&#10;" WIDTH="1616" HEIGHT="3128" HPOS="466" VPOS="276"></String>
+            <String CONTENT=" &#10;     &#10;    &#10;   &#10;  &#10; &#10;&#10;" WIDTH="1616" HEIGHT="3128" HPOS="466" VPOS="276"></String>
+            <String CONTENT=" &#10;     &#10;    &#10;   &#10;  &#10; &#10;&#10;" WIDTH="1616" HEIGHT="3128" HPOS="466" VPOS="276"></String>
+            <String CONTENT=" &#10;     &#10;    &#10;   &#10;  &#10; &#10;&#10;" WIDTH="1616" HEIGHT="3128" HPOS="466" VPOS="276"></String>
+            <String CONTENT=" &#10;     &#10;    &#10;   &#10;  &#10; &#10;&#10;" WIDTH="1616" HEIGHT="3128" HPOS="466" VPOS="276"></String>
+            <String CONTENT=" &#10;     &#10;    &#10;   &#10;  &#10; &#10;&#10;" WIDTH="1616" HEIGHT="3128" HPOS="466" VPOS="276"></String>
+          </TextLine>
+        </TextBlock>
+      </PrintSpace>
+    </Page>
+  </Layout>
+</alto>
diff --git a/spec/fixtures/derivatives/b6-thumbnail.jpeg b/spec/fixtures/derivatives/b6-thumbnail.jpeg
diff --git a/spec/indexers/generic_work_resource_indexer_spec.rb b/spec/indexers/generic_work_resource_indexer_spec.rb
@@ -7,7 +7,7 @@
 
 RSpec.describe GenericWorkResourceIndexer do
   let(:indexer_class) { described_class }
-  let(:resource) { GenericWorkResource.new }
+  let!(:resource) { Hyrax.persister.save(resource: GenericWorkResource.new) }
 
   it_behaves_like 'a Hyrax::Resource indexer'
 end
diff --git a/spec/indexers/image_resource_indexer_spec.rb b/spec/indexers/image_resource_indexer_spec.rb
@@ -7,7 +7,7 @@
 
 RSpec.describe ImageResourceIndexer do
   let(:indexer_class) { described_class }
-  let(:resource) { ImageResource.new }
+  let!(:resource) { Hyrax.persister.save(resource: ImageResource.new) }
 
   it_behaves_like 'a Hyrax::Resource indexer'
 end
diff --git a/spec/models/concerns/hyrax/solr_document_behavior_decorator_spec.rb b/spec/models/concerns/hyrax/solr_document_behavior_decorator_spec.rb