newspaper_works 0.1.0

data/spec/indexers/concerns/newspaper_works/indexes_relationships_spec.rb

@@ -0,0 +1,86 @@
+require 'spec_helper'
+require 'model_shared'
+
+RSpec.describe NewspaperWorks::IndexesRelationships do
+  # use an instance var so we can create fixtures only once
+  before(:all) { @page_for_indexrel, @page2 = model_fixtures(:newspaper_pages) }
+  let(:page_indexer) { NewspaperPageIndexer.new(@page_for_indexrel) }
+  let(:solr_doc) { {} }
+
+  describe '#index_relationships' do
+    before { page_indexer.index_relationships(@page_for_indexrel, solr_doc) }
+    it 'sets the relationship fields correctly' do
+      expect(solr_doc['publication_id_ssi']).not_to be_falsey
+      expect(solr_doc['container_id_ssi']).not_to be_falsey
+      expect(solr_doc['issue_id_ssi']).not_to be_falsey
+      expect(solr_doc['article_ids_ssim']).not_to be_falsey
+    end
+  end
+
+  describe '#index_publication' do
+    before { page_indexer.index_publication(@page_for_indexrel, solr_doc) }
+    it 'sets the publication title fields correctly' do
+      expect(solr_doc['publication_id_ssi']).not_to be_falsey
+      expect(solr_doc['publication_title_ssi']).to eq("Yesterday's News")
+      expect(solr_doc['publication_unique_id_ssi']).to eq("sn1234567")
+    end
+  end
+
+  describe '#index_container' do
+    before { page_indexer.index_container(@page_for_indexrel, solr_doc) }
+    it 'sets the container fields correctly' do
+      expect(solr_doc['container_id_ssi']).not_to be_falsey
+      expect(solr_doc['container_title_ssi']).to eq('Reel123a')
+    end
+  end
+
+  describe '#index_issue' do
+    before { page_indexer.index_issue(@page_for_indexrel, solr_doc) }
+    it 'sets the issue fields correctly' do
+      expect(solr_doc['issue_id_ssi']).not_to be_falsey
+      expect(solr_doc['issue_title_ssi']).to eq('December 7, 1941')
+      expect(solr_doc['issue_edition_number_ssi']).to eq('1')
+      expect(solr_doc['publication_date_dtsi']).to eq('1941-12-07T00:00:00Z')
+    end
+  end
+
+  describe '#index_pages' do
+    let(:article) { @page_for_indexrel.articles.first }
+    let(:article_indexer) { NewspaperArticleIndexer.new(article) }
+    before { page_indexer.index_pages(article, solr_doc) }
+    it 'sets the issue fields correctly' do
+      expect(solr_doc['page_ids_ssim'].first).to eq(@page_for_indexrel.id)
+      expect(solr_doc['page_titles_ssim'].first).to eq('Page 1')
+    end
+  end
+
+  describe '#index_siblings' do
+    let(:solr_doc_2) { {} }
+    before do
+      page_indexer.index_siblings(@page_for_indexrel, solr_doc)
+      page_indexer.index_siblings(@page2, solr_doc_2)
+    end
+    it 'sets the prev/next fields correctly' do
+      expect(solr_doc['is_preceding_page_of_ssi']).not_to be_falsey
+      expect(solr_doc['is_following_page_of_ssi']).to be_nil
+      expect(solr_doc['first_page_bsi']).to be_truthy
+      expect(solr_doc_2['is_preceding_page_of_ssi']).to be_nil
+      expect(solr_doc_2['is_following_page_of_ssi']).not_to be_falsey
+    end
+  end
+
+  describe '#index_articles' do
+    before { page_indexer.index_articles(@page_for_indexrel, solr_doc) }
+    it 'sets the article fields correctly' do
+      expect(solr_doc['article_ids_ssim']).not_to be_blank
+      expect(solr_doc['article_titles_ssim'].first).to eq('Happening now')
+    end
+  end
+
+  describe '#index_parent_facets' do
+    before { page_indexer.index_parent_facets(@page_for_indexrel.issue, solr_doc) }
+    it 'sets the facet fields correctly' do
+      expect(solr_doc['language_sim']).not_to be_blank
+    end
+  end
+end

data/spec/indexers/newspaper_article_indexer_spec.rb

@@ -0,0 +1,29 @@
+require 'spec_helper'
+
+RSpec.describe NewspaperArticleIndexer do
+  let(:article) do
+    NewspaperArticle.new(
+      id: 'foo1234',
+      title: ['Whatever'],
+      genre: %w[http://id.loc.gov/vocabulary/graphicMaterials/tgm000098 FOO]
+    )
+  end
+  let(:indexer) { described_class.new(article) }
+
+  describe '#generate_solr_document' do
+    subject { indexer.generate_solr_document }
+
+    it 'adds the correct fields to the Solr document' do
+      expect(subject['genre_tesim']).not_to be_falsey
+      expect(subject['genre_sim']).not_to be_falsey
+    end
+
+    it 'indexes genre terms with a URI correctly' do
+      expect(subject['genre_tesim']).to include('Advertisement')
+    end
+
+    it 'indexes genre terms without a URI correctly' do
+      expect(subject['genre_sim']).to include('FOO')
+    end
+  end
+end

data/spec/indexers/newspaper_issue_indexer_spec.rb

@@ -0,0 +1,19 @@
+require 'spec_helper'
+
+RSpec.describe NewspaperIssueIndexer do
+  let(:issue) do
+    NewspaperIssue.new(
+      id: 'foo1234',
+      title: ['Whatever']
+    )
+  end
+  let(:indexer) { described_class.new(issue) }
+
+  describe '#generate_solr_document' do
+    subject { indexer.generate_solr_document }
+
+    it 'adds the default edition field to the Solr document' do
+      expect(subject['edition_number_tesim']).to eq('1')
+    end
+  end
+end

data/spec/indexers/newspaper_title_indexer_spec.rb

@@ -0,0 +1,22 @@
+require 'spec_helper'
+
+RSpec.describe NewspaperTitleIndexer do
+  let(:ntitle) do
+    NewspaperTitle.new(
+      id: 'foo1234',
+      title: ['Whatever'],
+      publication_date_start: '1975',
+      publication_date_end: '1995'
+    )
+  end
+  let(:indexer) { described_class.new(ntitle) }
+
+  describe '#generate_solr_document' do
+    subject { indexer.generate_solr_document }
+
+    it 'indexes date ranges correctly' do
+      expect(subject['publication_date_start_dtsi']).to eq('1975-01-01T00:00:00Z')
+      expect(subject['publication_date_end_dtsi']).to eq('1995-12-31T23:59:59Z')
+    end
+  end
+end

data/spec/indexers/newspaper_works/newspaper_core_indexer_spec.rb

@@ -0,0 +1,23 @@
+require 'spec_helper'
+
+RSpec.describe NewspaperWorks::NewspaperCoreIndexer do
+  let(:geonames_uri) { 'http://sws.geonames.org/4950065/' }
+  let(:pop) { Hyrax::ControlledVocabularies::Location.new(geonames_uri) }
+  let(:article) do
+    NewspaperArticle.new(
+      id: 'foo1234',
+      title: ['Whatever'],
+      place_of_publication: pop
+    )
+  end
+  let(:indexer) { described_class.new(article) }
+
+  describe '#generate_solr_document' do
+    subject { indexer.generate_solr_document }
+    it 'processes place_of_publication field' do
+      expect(subject['place_of_publication_tesim']).to include(geonames_uri)
+      expect(subject['place_of_publication_city_sim']).to include('Salem')
+      expect(subject['place_of_publication_state_sim']).to include('Massachusetts')
+    end
+  end
+end

data/spec/lib/newspaper_works/configuration_spec.rb

@@ -0,0 +1,18 @@
+require 'spec_helper'
+
+RSpec.describe NewspaperWorks::Configuration do
+  subject { described_class.new }
+
+  it { is_expected.to respond_to(:publication_unique_id_property) }
+  it { is_expected.to respond_to(:publication_unique_id_field) }
+
+  describe '#publicationunique_id_property' do
+    subject { described_class.new.publication_unique_id_property }
+    it { is_expected.to eq(:lccn) }
+  end
+
+  describe '#publication_unique_id_field' do
+    subject { described_class.new.publication_unique_id_field }
+    it { is_expected.to eq('lccn_tesim') }
+  end
+end

data/spec/lib/newspaper_works/data/work_derivatives_spec.rb

@@ -0,0 +1,245 @@
+# encoding: UTF-8
+
+require 'spec_helper'
+require 'misc_shared'
+
+RSpec.describe NewspaperWorks::Data::WorkDerivatives do
+  include_context "shared setup"
+
+  let(:bare_work) do
+    work = NewspaperPage.new
+    work.title = ['Another one']
+    work.save!
+    work
+  end
+
+  let(:work) do
+    # sample work comes from shared setup, but we need derivative, save...
+    mk_txt_derivative(sample_work)
+    sample_work.save!(validate: false)
+    sample_work
+  end
+
+  let(:adapter) { described_class.new(work) }
+
+  let(:txt1) do
+    whitelist = Hyrax.config.whitelisted_ingest_dirs
+    whitelist.push('/tmp') unless whitelist.include?('/tmp')
+    file = Tempfile.new(['txt1', '.txt'])
+    file.write('hello')
+    file.flush
+    file
+  end
+
+  let(:txt2) do
+    file = Tempfile.new('txt2.txt')
+    file.write('bye')
+    file.flush
+    file
+  end
+
+  let(:encoded_text) do
+    file = Tempfile.new('txt_encoded.txt', encoding: 'UTF-8')
+    file.write('Gorgonzola Dolce® — on sale for £12.50/kg')
+    file.flush
+    file
+  end
+
+  describe "enumerates available derivatives like hash" do
+    it "includes expected derivative path for work" do
+      expect(adapter.keys).to include 'txt'
+    end
+
+    it "can be introspected for quantity of derivatives" do
+      # `size` method without argument is count of derivatives,
+      #   functions equivalently to adapter.keys.size
+      expect(adapter.size).to eq adapter.keys.size
+    end
+
+    it "enumerates expected derivative extension for work" do
+      ext_found = adapter.keys
+      expect(ext_found).to include 'txt'
+    end
+
+    it "enumerates expected derivative extension for file set" do
+      file_set = work.members.select { |m| m.class == FileSet }[0]
+      adapter = described_class.new(file_set)
+      ext_found = adapter.keys
+      expect(ext_found).to include 'txt'
+    end
+
+    it "enumerates expected derivative extension for file set id" do
+      file_set = work.members.select { |m| m.class == FileSet }[0]
+      adapter = described_class.new(file_set.id)
+      ext_found = adapter.keys
+      expect(ext_found).to include 'txt'
+    end
+  end
+
+  describe "loads derivatives for a work" do
+    it "Loads text derivative path" do
+      expect(File.exist?(adapter.path('txt'))).to be true
+      expect(adapter.exist?('txt')).to be true
+    end
+
+    it "Loads text derivative data" do
+      expect(adapter.data('txt')).to include 'mythical'
+    end
+
+    it "Handles character encoding on read" do
+      # replace fixture text derivative for work with encoded text
+      adapter.attach(encoded_text.path, 'txt')
+      data = adapter.data('txt')
+      expect(data).to include '—' # em-dash
+      expect(data).to include '£' # gb-pound sign
+      expect(data.encoding.to_s).to eq 'UTF-8'
+    end
+
+    it "Loads thumbnail derivative data" do
+      mk_thumbnail_derivative(work)
+      # get size by loading data
+      expect(adapter.data('thumbnail').bytes.size).to eq 16_743
+      # get size by File.size via .size method
+      expect(adapter.size('thumbnail')).to eq 16_743
+    end
+
+    it "Can access jp2 derivative" do
+      mk_jp2_derivative(work)
+      expect(File.exist?(adapter.path('jp2'))).to be true
+      expect(adapter.exist?('jp2')).to be true
+    end
+  end
+
+  describe "create, update, delete derivatives" do
+    it "will queue derivative file assignment" do
+      adapter = described_class.new(bare_work)
+      adapter.assign(example_gray_jp2)
+      expect(adapter.assigned).to include example_gray_jp2
+    end
+
+    it "will fail to assign file in non-whitelisted dir" do
+      adapter = described_class.new(bare_work)
+      # need a non-whitlisted file that exists:
+      bad_path = File.expand_path("../../spec_helper.rb", fixture_path)
+      expect { adapter.assign(bad_path) }.to raise_error(SecurityError)
+    end
+
+    it "will remove file assignment from queue" do
+      adapter = described_class.new(bare_work)
+      expect(adapter.state).to eq 'empty'
+      adapter.assign(example_gray_jp2)
+      expect(adapter.assigned).to include example_gray_jp2
+      expect(adapter.state).to eq 'dirty'
+      adapter.unassign(example_gray_jp2)
+      expect(adapter.assigned).not_to include example_gray_jp2
+      expect(adapter.state).to eq 'empty'
+    end
+
+    it "will queue a deletion" do
+      # Given a work with a derivative (txt) already assigned
+      expect(adapter.state).to eq 'saved'
+      # unassigning path...
+      adapter.unassign('txt')
+      # will lead to queued unassignment (intent to delete)...
+      expect(adapter.unassigned).to include 'txt'
+      # and a 'dirty' adapter state (unflushed changes):
+      expect(adapter.state).to eq 'dirty'
+    end
+
+    it "will flush a removal and addition on commit!" do
+      # Given a work with a derivative (txt) already assigned
+      expect(adapter.keys).to include 'txt'
+      expect(adapter.keys).not_to include 'jp2'
+      # unassigning path...
+      adapter.unassign('txt')
+      # and assigning another attachment:
+      adapter.assign(example_gray_jp2)
+      # ...committing these will flush the changes (synchronously):
+      adapter.commit!
+      expect(adapter.keys).not_to include 'txt'
+      expect(adapter.keys).to include 'jp2'
+      expect(adapter.size('jp2')).to eq 27_703
+    end
+
+    it "can attach derivative from file" do
+      expect(adapter.keys).not_to include 'jp2'
+      adapter.attach(example_gray_jp2, 'jp2')
+      expect(adapter.exist?('jp2')).to be true
+      expect(adapter.path('jp2')).not_to be nil
+      expect(File.size(adapter.path('jp2'))).to eq File.size(example_gray_jp2)
+      expect(adapter.keys).to include 'jp2'
+      d_path = path_factory.derivative_path_for_reference(adapter.fileset_id, 'jp2')
+      expect(adapter.values).to include d_path
+    end
+
+    it "can replace aderivative with new attachment" do
+      adapter.attach(txt1.path, 'txt')
+      expect(adapter.data('txt')).to eq 'hello'
+      adapter.attach(txt2.path, 'txt')
+      expect(adapter.data('txt')).to eq 'bye'
+    end
+
+    it "can delete an attached derivative" do
+      adapter.attach(txt1.path, 'txt')
+      expect(adapter.keys).to include 'txt'
+      expect(adapter.data('txt')).to eq 'hello'
+      adapter.delete('txt')
+      expect(adapter.path('txt')).to be nil
+      expect(adapter.keys).not_to include 'txt'
+    end
+
+    it "persists log of attachment to RDBMS" do
+      adapter.assign(txt1.path)
+      result = NewspaperWorks::DerivativeAttachment.find_by(
+        fileset_id: adapter.fileset.id,
+        path: txt1.path,
+        destination_name: 'txt'
+      )
+      expect(result).not_to be_nil
+    end
+
+    it "persists a log of path relation to primary file" do
+      # this is an integration test by practical necessity, with
+      #   WorkFiles adapting a bare work with no fileset.
+      work_files = NewspaperWorks::Data::WorkFiles.of(bare_work)
+      work_files.assign(example_gray_jp2)
+      adapter = work_files.derivatives
+      adapter.assign(txt1.path)
+      result = NewspaperWorks::IngestFileRelation.find_by(
+        derivative_path: txt1.path,
+        file_path: example_gray_jp2
+      )
+      expect(result).not_to be_nil
+    end
+
+    it "commits queued derivatives" do
+      NewspaperWorks::IngestFileRelation.where(file_path: example_gray_jp2).delete_all
+      work_files = NewspaperWorks::Data::WorkFiles.of(bare_work)
+      work_files.assign(example_gray_jp2)
+      adapter = work_files.derivatives
+      adapter.assign(txt1.path)
+      expect(File.exist?(txt1.path)).to be true
+      expect(adapter.keys.size).to eq 0
+      # we need a fileset, saved with import_url, attached to work:
+      fileset = valid_file_set
+      fileset.import_url = 'file://' + example_gray_jp2
+      fileset.save!
+      bare_work.members.push(fileset)
+      bare_work.save!
+      fileset.reload
+      expect(fileset.member_of[0].id).to eq bare_work.id
+      # with a new adapter instance...
+      adapter2 = described_class.of(bare_work)
+      # call .commit_queued! with our fileset...
+      expect(File.exist?(txt1.path)).to be true
+      adapter2.commit_queued!(fileset)
+      # ...which should result in saved, reloaded derivative...
+      expect(adapter2.keys.size).to eq 1
+      expect(File.size(adapter2.values[0])).to eq File.size(txt1.path)
+      # ...also found via Hyrax::DerviativePath:
+      found = Hyrax::DerivativePath.derivatives_for_reference(fileset.id)
+      expect(found.size).to eq 1
+      expect(File.size(found[0])).to eq File.size(txt1.path)
+    end
+  end
+end

data/spec/lib/newspaper_works/data/work_file_spec.rb

@@ -0,0 +1,99 @@
+require 'spec_helper'
+require 'misc_shared'
+
+RSpec.describe NewspaperWorks::Data::WorkFile do
+  include_context "shared setup"
+
+  # sample objects:
+  let(:work) { work_with_file }
+
+  describe "adapter composition" do
+    it "adapts work with nil fileset" do
+      adapter = described_class.new(work)
+      expect(adapter.work).to be work
+      expect(adapter.fileset).to be_nil
+    end
+
+    it "adapts work with 'of' alt constructor" do
+      adapter = described_class.of(work)
+      expect(adapter.work).to be work
+    end
+
+    it "adapts work and explicitly provided fileset" do
+      fileset = work.members.select { |m| m.class == FileSet }[0]
+      adapter = described_class.of(work, fileset)
+      expect(adapter.work).to be work
+      expect(adapter.fileset).to be fileset
+    end
+
+    it "constructs with a parent object, if provided" do
+      fileset = work.members.select { |m| m.class == FileSet }[0]
+      parent = double('parent')
+      adapter = described_class.of(work, fileset, parent)
+      expect(adapter.parent).to be parent
+    end
+  end
+
+  describe "read file metadata" do
+    it "gets original filename" do
+      fileset = work.members.select { |m| m.class == FileSet }[0]
+      adapter = described_class.of(work, fileset)
+      expect(adapter.name).to eq fileset.original_file.original_name
+      expect(adapter.name).to eq 'credits.md'
+    end
+
+    it "gets miscellaneous metadata field values" do
+      fileset = work.members.select { |m| m.class == FileSet }[0]
+      adapter = described_class.of(work, fileset)
+      # expectations for accessors of size, date_*, mime_type
+      expect(adapter.size).to eq File.size(txt_path)
+      expect(adapter.name).to eq 'credits.md'
+      expect(adapter.mime_type).to eq 'text/plain'
+      # getting actual value for date fields requires digging through
+      #   multiple layers of ActiveTuples indirection...
+      expect(adapter.date_created.to_a[0].to_s).to eq static_date.to_s
+      expect(adapter.date_modified.to_a[0].to_s).to eq static_date.to_s
+    end
+  end
+
+  describe "read binary via transparent repository checkout" do
+    it "gets path (from checkout)" do
+      fileset = work.members.select { |m| m.class == FileSet }[0]
+      adapter = described_class.of(work, fileset)
+      # Get a path to a working copy
+      path = adapter.path
+      expect(path).to be_a String
+      expect(File.exist?(path)).to be true
+      # size of working copy binary checkout matches size in computed metadata
+      expect(File.size(path)).to eq fileset.original_file.size
+    end
+
+    it "gets data as bytes" do
+      fileset = work.members.select { |m| m.class == FileSet }[0]
+      adapter = described_class.of(work, fileset)
+      # Get a data from the working copy
+      data = adapter.data
+      expect(data).to be_a String
+      # size of working copy binary checkout matches size in computed metadata
+      expect(data.size).to eq fileset.original_file.size
+    end
+
+    it "runs block on data as IO" do
+      fileset = work.members.select { |m| m.class == FileSet }[0]
+      adapter = described_class.of(work, fileset)
+      adapter.with_io { |io| expect(io.read.size).to eq File.size(txt_path) }
+    end
+  end
+
+  describe "derivative access" do
+    it "gets derivatives for file" do
+      fileset = work.members.select { |m| m.class == FileSet }[0]
+      adapter = described_class.of(work, fileset)
+      expect(adapter.derivatives.class).to eq \
+        NewspaperWorks::Data::WorkDerivatives
+      expect(adapter.derivatives.fileset).to be fileset
+      expect(adapter.derivatives.work).to be work
+      expect(adapter.derivatives.parent).to be adapter
+    end
+  end
+end