ro-crate 0.4.11 → 0.4.12

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 62f3979b325743d31720f803dbf75690ad6b7b9bcf9f2cbf63d8200a2bb204ba
-  data.tar.gz: 3a229e06492a3f9a90721799f43929d6a5ca01fc0a0ff34d917eeebff037d9d4
+  metadata.gz: d9f081110d1e5b5e94674cb53d07262fbce50ec3f63437d5c4c1ecda8b04f835
+  data.tar.gz: 0cd710a9cd86063e706fd9cf338e5fdf55f5bc6993f306af4f50d6f9191a0d87
 SHA512:
-  metadata.gz: 242b8e326756d51ab583726db7fb94763286f2764bdf1f0523cd9dab5fa560fe4da3016a4a4b84b0cc6bcef4ec096aeae7dc2fd344ef6cb5ee045cf99a954ef9
-  data.tar.gz: 4abaece91d997ac398ee627f639fc98c924768c255bac1bb6c9ce7f43fc03f5b018a2fc657008863c65537592261e85f697ef2c7e200c20621ba41ff0a933541
+  metadata.gz: 8ce573fbd259108edbb528e24f0d9647cfee4454378d57a4b3392aaeb0dea80385b2828221a0979caa4733be0e2108a4a535725846f576c9ac97866f46911e47
+  data.tar.gz: ffc7aafbdccb3f4897ad2acae726714d7a313f02fd87e5c96173da2b12860dfa1318b9b323dde751e3ac10c9a66cf29effea383037bc4622eaf5113dd01bd426

data/Gemfile.lock

@@ -1,21 +1,21 @@
 PATH
   remote: .
   specs:
-    ro-crate (0.4.11)
-      addressable (~> 2.7.0)
+    ro-crate (0.4.12)
+      addressable (>= 2.7, < 2.9)
       rubyzip (~> 2.0.0)
 
 GEM
   remote: https://rubygems.org/
   specs:
-    addressable (2.7.0)
+    addressable (2.8.0)
       public_suffix (>= 2.0.2, < 5.0)
     crack (0.4.3)
       safe_yaml (~> 1.0.0)
     docile (1.3.5)
     hashdiff (1.0.1)
-    power_assert (0.4.1)
-    public_suffix (4.0.3)
+    power_assert (1.1.3)
+    public_suffix (4.0.6)
     rake (13.0.0)
     rubyzip (2.0.0)
     safe_yaml (1.0.5)
@@ -25,7 +25,7 @@ GEM
       simplecov_json_formatter (~> 0.1)
     simplecov-html (0.12.3)
     simplecov_json_formatter (0.1.2)
-    test-unit (3.2.3)
+    test-unit (3.2.9)
       power_assert
     webmock (3.8.3)
       addressable (>= 2.3.6)

data/README.md

@@ -1,5 +1,7 @@
 # ro-crate-ruby
 
+![Tests](https://github.com/ResearchObject/ro-crate-ruby/actions/workflows/tests.yml/badge.svg)
+
 This is a WIP gem for creating, manipulating and reading RO-Crates (conforming to version 1.1 of the specification).
 
 * RO-Crate - https://researchobject.github.io/ro-crate/

data/lib/ro_crate/model/crate.rb

@@ -25,6 +25,15 @@ module ROCrate
       super(self, nil, id, properties)
     end
 
+    ##
+    # Lookup an Entity using the given ID (in this Entity's crate).
+    #
+    # @param id [String] The ID to query.
+    # @return [Entity, nil]
+    def dereference(id)
+      entities.detect { |e| e.canonical_id == crate.resolve_id(id) } if id
+    end
+
     ##
     # Create a new file and add it to the crate.
     #
@@ -229,21 +238,21 @@ module ROCrate
       entity.class.new(crate, entity.id, entity.raw_properties)
     end
 
-    alias_method :own_entries, :entries
+    alias_method :own_payload, :payload
     ##
-    # # The RO-Crate's "payload" of the crate - a map of all the files/directories contained in the RO-Crate, where the
-    # key is the destination path within the crate and the value is an Entry where the source data can be read.
+    # The file payload of the RO-Crate - a map of all the files/directories contained in the RO-Crate, where the
+    # key is the path relative to the crate's root, and the value is an Entry where the source data can be read.
     #
     # @return [Hash{String => Entry}>]
-    def entries
+    def payload
       # Gather a map of entries, starting from the crate itself, then any directory data entities, then finally any
       # file data entities. This ensures in the case of a conflict, the more "specific" data entities take priority.
-      entries = own_entries
+      entries = own_payload
       non_self_entities = default_entities.reject { |e| e == self }
       sorted_entities = (non_self_entities | data_entities).sort_by { |e| e.is_a?(ROCrate::Directory) ? 0 : 1 }
 
       sorted_entities.each do |entity|
-        entity.entries.each do |path, entry|
+        entity.payload.each do |path, entry|
           entries[path] = entry
         end
       end
@@ -255,6 +264,29 @@ module ROCrate
       binding
     end
 
+    ##
+    # Remove the entity from the RO-Crate.
+    #
+    # @param entity [Entity, String] The entity or ID of an entity to remove from the crate.
+    # @param remove_orphaned [Boolean] Should linked contextual entities also be removed from the crate they are left
+    #                                   dangling (nothing else is linked to them)?
+    #
+    # @return [Entity, nil] The entity that was deleted, or nil if nothing was deleted.
+    def delete(entity, remove_orphaned: true)
+      entity = dereference(entity) if entity.is_a?(String)
+      return unless entity
+
+      deleted = data_entities.delete(entity) || contextual_entities.delete(entity)
+
+      if deleted && remove_orphaned
+        crate_entities = crate.linked_entities(deep: true)
+        to_remove = (entity.linked_entities(deep: true) - crate_entities)
+        to_remove.each(&:delete)
+      end
+
+      deleted
+    end
+
     private
 
     def full_entry_path(relative_path)

data/lib/ro_crate/model/data_entity.rb

@@ -24,12 +24,13 @@ module ROCrate
     end
 
     ##
-    # A map of all the files/directories associated with this DataEntity.
+    # The payload of all the files/directories associated with this DataEntity, mapped by their relative file path.
     #
     # @return [Hash{String => Entry}>] The key is the location within the crate, and the value is an Entry.
-    def entries
+    def payload
       {}
     end
+    alias_method :entries, :payload
 
     ##
     # A disk-safe filepath based on the ID of this DataEntity.

data/lib/ro_crate/model/directory.rb

@@ -28,11 +28,11 @@ module ROCrate
     end
 
     ##
-    # The "payload" of this directory - a map of all the files/directories, where the key is the destination path
+    # The payload of this directory - a map of all the files/directories, where the key is the destination path
     # within the crate and the value is an Entry where the source data can be read.
     #
     # @return [Hash{String => Entry}>]
-    def entries
+    def payload
       entries = {}
       entries[filepath.chomp('/')] = @entry if @entry
 

data/lib/ro_crate/model/entity.rb

@@ -129,16 +129,26 @@ module ROCrate
     # @param id [String] The ID to query.
     # @return [Entity, nil]
     def dereference(id)
-      crate.entities.detect { |e| e.canonical_id == crate.resolve_id(id) } if id
+      crate.dereference(id)
     end
-
     alias_method :get, :dereference
 
+    ##
+    # Remove this entity from the RO-Crate.
+    #
+    # @param remove_orphaned [Boolean] Should linked contextual entities also be removed from the crate (if nothing else is linked to them)?
+    #
+    # @return [Entity, nil] This entity, or nil if nothing was deleted.
+    def delete(remove_orphaned: true)
+      crate.delete(self, remove_orphaned: remove_orphaned)
+    end
+
     def id
       @properties['@id']
     end
 
     def id=(id)
+      @canonical_id = nil
       @properties['@id'] = self.class.format_id(id)
     end
 
@@ -190,13 +200,13 @@ module ROCrate
     #
     # @return [Addressable::URI]
     def canonical_id
-      crate.resolve_id(id)
+      @canonical_id ||= crate.resolve_id(id)
     end
 
     ##
     # Is this entity local to the crate or an external reference?
     #
-    # @return [boolean]
+    # @return [Boolean]
     def external?
       crate.canonical_id.host != canonical_id.host
     end
@@ -226,6 +236,33 @@ module ROCrate
       @properties.has_type?(type)
     end
 
+    ##
+    # Gather a list of entities linked to this one through its properties.
+    # @param deep [Boolean] If false, only consider direct links, otherwise consider transitive links.
+    # @param linked [Hash{String => Entity}] Discovered entities, mapped by their ID, to avoid loops when recursing.
+    # @return [Array<Entity>]
+    def linked_entities(deep: false, linked: {})
+      properties.each do |key, value|
+        value = [value] if value.is_a?(JSONLDHash)
+
+        if value.is_a?(Array)
+          value.each do |v|
+            if v.is_a?(JSONLDHash) && !linked.key?(v['@id'])
+              entity = v.dereference
+              linked[entity.id] = entity if entity
+              if deep
+                entity.linked_entities(deep: true, linked: linked).each do |e|
+                  linked[e.id] = e
+                end
+              end
+            end
+          end
+        end
+      end
+
+      linked.values.compact
+    end
+
     private
 
     def default_properties

data/lib/ro_crate/model/entry.rb

@@ -14,10 +14,10 @@ module ROCrate
     end
 
     ##
-    # Write the source to the destination via a buffer.
+    # Write the entry's source to the destination via a buffer.
     #
     # @param dest [#write] An IO-like destination to write to.
-    def write(dest)
+    def write_to(dest)
       input = source
       input = input.open('rb') if input.is_a?(Pathname)
       while (buff = input.read(4096))

data/lib/ro_crate/model/file.rb

@@ -56,7 +56,7 @@ module ROCrate
     # (for compatibility with Directory#entries)
     #
     # @return [Hash{String => Entry}>] The key is the location within the crate, and the value is an Entry.
-    def entries
+    def payload
       remote? ? {} : { filepath => source }
     end
 

data/lib/ro_crate/model/remote_entry.rb

@@ -2,7 +2,7 @@ module ROCrate
   ##
   # A class to represent a reference within an RO-Crate, to a remote file held on the internet somewhere.
   # It handles the actual reading/writing of bytes.
-  class RemoteEntry
+  class RemoteEntry < Entry
     attr_reader :uri
 
     ##
@@ -13,17 +13,6 @@ module ROCrate
       @uri = uri
     end
 
-    def write(dest)
-      raise 'Cannot write to a remote entry!'
-    end
-
-    ##
-    # Read from the source.
-    #
-    def read
-      source.read
-    end
-
     ##
     # @return [IO] An IO object for the remote resource.
     #

data/lib/ro_crate/reader.rb

@@ -181,9 +181,10 @@ module ROCrate
         crate.metadata.properties = entity_hash.delete(ROCrate::Metadata::IDENTIFIER)
         crate.metadata.context = context
         preview_properties = entity_hash.delete(ROCrate::Preview::IDENTIFIER)
-        if preview_properties
-          preview_path = ::File.join(source, ROCrate::Preview::IDENTIFIER)
-          crate.preview = ROCrate::Preview.new(crate, ::File.exists?(preview_path) ? Pathname.new(preview_path) : nil, preview_properties)
+        preview_path = ::File.join(source, ROCrate::Preview::IDENTIFIER)
+        preview_path = ::File.exists?(preview_path) ? Pathname.new(preview_path) : nil
+        if preview_properties || preview_path
+          crate.preview = ROCrate::Preview.new(crate, preview_path, preview_properties || {})
         end
         crate.add_all(source, false)
       end

data/lib/ro_crate/writer.rb

@@ -16,14 +16,14 @@ module ROCrate
     # @param overwrite [Boolean] Whether or not to overwrite existing files.
     def write(dir, overwrite: true)
       FileUtils.mkdir_p(dir) # Make any parent directories
-      @crate.entries.each do |path, entry|
+      @crate.payload.each do |path, entry|
         fullpath = ::File.join(dir, path)
         next if !overwrite && ::File.exist?(fullpath)
         next if entry.directory?
         FileUtils.mkdir_p(::File.dirname(fullpath))
         temp = Tempfile.new('ro-crate-temp')
         begin
-          entry.write(temp)
+          entry.write_to(temp)
           temp.close
           FileUtils.mv(temp, fullpath)
         ensure
@@ -38,9 +38,9 @@ module ROCrate
     # @param destination [String, ::File] The destination where to write the RO-Crate zip.
     def write_zip(destination)
       Zip::File.open(destination, Zip::File::CREATE) do |zip|
-        @crate.entries.each do |path, entry|
+        @crate.payload.each do |path, entry|
           next if entry.directory?
-          zip.get_output_stream(path) { |s| entry.write(s) }
+          zip.get_output_stream(path) { |s| entry.write_to(s) }
         end
       end
     end

data/lib/ro_crate.rb

@@ -10,8 +10,8 @@ require 'ro_crate/json_ld_hash'
 require 'ro_crate/model/entity'
 require 'ro_crate/model/data_entity'
 require 'ro_crate/model/file'
-require 'ro_crate/model/remote_entry'
 require 'ro_crate/model/entry'
+require 'ro_crate/model/remote_entry'
 require 'ro_crate/model/directory'
 require 'ro_crate/model/metadata'
 require 'ro_crate/model/preview_generator'

data/ro_crate.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name        = 'ro-crate'
-  s.version     = '0.4.11'
+  s.version     = '0.4.12'
   s.summary     = 'Create, manipulate, read RO-Crates.'
   s.authors     = ['Finn Bacall']
   s.email       = 'finn.bacall@manchester.ac.uk'
@@ -8,7 +8,7 @@ Gem::Specification.new do |s|
   s.homepage    = 'https://github.com/ResearchObject/ro-crate-ruby'
   s.require_paths = ['lib']
   s.licenses    = ['MIT']
-  s.add_runtime_dependency 'addressable', '~> 2.7.0'
+  s.add_runtime_dependency 'addressable', '>= 2.7', '< 2.9'
   s.add_runtime_dependency 'rubyzip', '~> 2.0.0'
   s.add_development_dependency 'rake', '~> 13.0.0'
   s.add_development_dependency 'test-unit', '~> 3.2.3'

data/test/crate_test.rb

@@ -215,7 +215,7 @@ class CrateTest < Test::Unit::TestCase
     crate = ROCrate::Crate.new
     entities = crate.add_all(fixture_file('directory').path, include_hidden: true)
 
-    paths = crate.entries.keys
+    paths = crate.payload.keys
     assert_equal 11, paths.length
     assert_includes paths, 'data'
     assert_includes paths, 'root.txt'
@@ -250,7 +250,7 @@ class CrateTest < Test::Unit::TestCase
 
     assert_empty entities
 
-    paths = crate.entries.keys
+    paths = crate.payload.keys
     assert_equal 11, paths.length
     assert_includes paths, 'data'
     assert_includes paths, 'root.txt'
@@ -278,7 +278,7 @@ class CrateTest < Test::Unit::TestCase
     crate = ROCrate::Crate.new
     entities = crate.add_all(fixture_file('directory').path)
 
-    paths = crate.entries.keys
+    paths = crate.payload.keys
     assert_equal 8, paths.length
     assert_includes paths, 'data'
     assert_includes paths, 'root.txt'
@@ -305,4 +305,26 @@ class CrateTest < Test::Unit::TestCase
 
     assert_equal "5678\n", crate.dereference('data/info.txt').source.read
   end
+
+  test 'can delete entities' do
+    crate = ROCrate::Crate.new
+    file = crate.add_file(StringIO.new(''), 'file')
+    person = crate.add_person('#bob', { name: 'Bob' })
+    file.author = person
+
+    assert crate.delete(file)
+    assert_not_include crate.entities, file
+    assert_not_include crate.entities, person
+  end
+
+  test 'can delete entities by id' do
+    crate = ROCrate::Crate.new
+    file = crate.add_file(StringIO.new(''), 'file')
+    person = crate.add_person('#bob', { name: 'Bob' })
+    file.author = person
+
+    assert crate.delete('file')
+    assert_not_include crate.entities, file
+    assert_not_include crate.entities, person
+  end
 end

data/test/directory_test.rb

@@ -5,41 +5,41 @@ class DirectoryTest < Test::Unit::TestCase
     crate = ROCrate::Crate.new
     crate.add_directory(fixture_file('directory'))
 
-    entries = crate.entries
+    payload = crate.payload
     base_path = ::File.dirname(fixture_file('directory'))
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/info.txt')), entries['directory/info.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/root.txt')), entries['directory/root.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data')), entries['directory/data'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/info.txt')), entries['directory/data/info.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/nested.txt')), entries['directory/data/nested.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/binary.jpg')), entries['directory/data/binary.jpg'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/info.txt')), payload['directory/info.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/root.txt')), payload['directory/root.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data')), payload['directory/data'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/info.txt')), payload['directory/data/info.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/nested.txt')), payload['directory/data/nested.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/binary.jpg')), payload['directory/data/binary.jpg'].path
   end
 
   test 'adding directory via path' do
     crate = ROCrate::Crate.new
     crate.add_directory(fixture_file('directory').path.to_s)
 
-    entries = crate.entries
+    payload = crate.payload
     base_path = ::File.dirname(fixture_file('directory'))
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/info.txt')), entries['directory/info.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/root.txt')), entries['directory/root.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data')), entries['directory/data'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/info.txt')), entries['directory/data/info.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/nested.txt')), entries['directory/data/nested.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/binary.jpg')), entries['directory/data/binary.jpg'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/info.txt')), payload['directory/info.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/root.txt')), payload['directory/root.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data')), payload['directory/data'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/info.txt')), payload['directory/data/info.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/nested.txt')), payload['directory/data/nested.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/binary.jpg')), payload['directory/data/binary.jpg'].path
   end
 
   test 'adding to given path' do
     crate = ROCrate::Crate.new
     crate.add_directory(fixture_file('directory').path.to_s, 'fish')
 
-    entries = crate.entries
+    payload = crate.payload
     base_path = ::File.dirname(fixture_file('directory'))
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/info.txt')), entries['fish/info.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/root.txt')), entries['fish/root.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data')), entries['fish/data'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/info.txt')), entries['fish/data/info.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/nested.txt')), entries['fish/data/nested.txt'].path
-    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/binary.jpg')), entries['fish/data/binary.jpg'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/info.txt')), payload['fish/info.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/root.txt')), payload['fish/root.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data')), payload['fish/data'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/info.txt')), payload['fish/data/info.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/nested.txt')), payload['fish/data/nested.txt'].path
+    assert_equal ::File.expand_path(::File.join(base_path, 'directory/data/binary.jpg')), payload['fish/data/binary.jpg'].path
   end
 end

data/test/entity_test.rb

@@ -91,4 +91,118 @@ class EntityTest < Test::Unit::TestCase
     assert_equal "http://www.data.com/my%20crate/", ROCrate::Crate.format_id('http://www.data.com/my%20crate'), 'Crate ID should end with /'
     assert_equal "http://www.data.com/my%20crate/", ROCrate::Crate.format_id('http://www.data.com/my%20crate/')
   end
+
+  test 'linked entities' do
+    crate = ROCrate::Crate.new
+    file = crate.add_file(StringIO.new(''), 'file')
+    person = crate.add_person('#bob', { name: 'Bob' })
+    file.author = person
+
+    linked = file.linked_entities
+    assert_include linked, person
+    assert_equal 1, linked.length
+
+    linked = crate.linked_entities
+    assert_include linked, file
+    assert_equal 1, linked.length
+
+    linked = crate.linked_entities(deep: true)
+    assert_include linked, file
+    assert_include linked, person
+    assert_equal 2, linked.length
+  end
+
+  test 'deleting entities removes linked entities' do
+    crate = ROCrate::Crate.new
+    file = crate.add_file(StringIO.new(''), 'file')
+    person = crate.add_person('#bob', { name: 'Bob' })
+    file.author = person
+
+    assert file.delete
+    assert_not_include crate.entities, file
+    assert_not_include crate.entities, person
+  end
+
+  test 'deleting entities does not remove dangling entities if option set' do
+    crate = ROCrate::Crate.new
+    file = crate.add_file(StringIO.new(''), 'file')
+    person = crate.add_person('#bob', { name: 'Bob' })
+    file.author = person
+
+    assert file.delete(remove_orphaned: false)
+    assert_not_include crate.entities, file
+    assert_include crate.entities, person
+  end
+
+  test 'creating files in various ways' do
+    stub_request(:get, 'http://example.com/external_ref.txt').to_return(status: 200, body: 'file contents')
+
+    crate = ROCrate::Crate.new
+
+    f1 = nil
+    Dir.chdir(fixture_dir) { f1 = ROCrate::File.new(crate, 'data.csv') }
+    refute f1.source.remote?
+    refute f1.source.directory?
+    assert_equal 20, f1.source.read.length
+    t = Tempfile.new('tf1')
+    f1.source.write_to(t)
+    t.rewind
+    assert_equal 20, t.read.length
+
+    f2 = ROCrate::File.new(crate, fixture_file('info.txt'), { author: crate.add_person('bob', name: 'Bob').reference })
+    refute f2.source.remote?
+    refute f2.source.directory?
+    assert f2.source.read
+    assert_equal 6, f2.source.read.length
+    t = Tempfile.new('tf2')
+    f2.source.write_to(t)
+    t.rewind
+    assert_equal 6, t.read.length
+
+    f3 = ROCrate::File.new(crate, 'http://example.com/external_ref.txt')
+    assert f3.source.remote?
+    refute f3.source.directory?
+    assert f3.source.read
+    assert_equal 13, f3.source.read.length
+    t = Tempfile.new('tf3')
+    f3.source.write_to(t)
+    t.rewind
+    assert_equal 13, t.read.length
+
+    f3 = ROCrate::File.new(crate, 'http://example.com/external_ref.txt')
+    assert f3.source.remote?
+    refute f3.source.directory?
+    assert f3.source.read
+    assert_equal 13, f3.source.read.length
+    t = Tempfile.new('tf3')
+    f3.source.write_to(t)
+    t.rewind
+    assert_equal 13, t.read.length
+  end
+
+  test 'assigning and checking type' do
+    crate = ROCrate::Crate.new
+    file = ROCrate::File.new(crate, 'data.csv')
+
+    assert file.has_type?('File')
+
+    file.type = ['File', 'Workflow']
+
+    assert file.has_type?('Workflow')
+    assert file.has_type?('File')
+    refute file.has_type?('Banana')
+  end
+
+  test 'inspecting object truncates very long property list' do
+    crate = ROCrate::Crate.new
+    entity = ROCrate::ContextualEntity.new(crate, 'hello')
+
+    assert entity.inspect.start_with?("<#ROCrate::ContextualEntity arcp://uuid,")
+
+    entity['veryLong'] = ('123456789a' * 100)
+
+    ins = entity.inspect
+    assert ins.length < 1000
+    assert ins.end_with?('...>')
+  end
 end

data/test/fixtures/conflicting_data_directory/info.txt

@@ -0,0 +1 @@
+abcd

data/test/fixtures/conflicting_data_directory/nested.txt

@@ -0,0 +1 @@
+No, I am nested!

data/test/reader_test.rb

@@ -88,12 +88,12 @@ class ReaderTest < Test::Unit::TestCase
   test 'reading from zip with directories' do
     crate = ROCrate::Reader.read_zip(fixture_file('directory.zip'))
 
-    assert crate.entries['fish/info.txt']
-    assert_equal '1234', crate.entries['fish/info.txt'].source.read.chomp
-    assert crate.entries['fish/root.txt']
-    assert crate.entries['fish/data/info.txt']
-    assert crate.entries['fish/data/nested.txt']
-    assert crate.entries['fish/data/binary.jpg']
+    assert crate.payload['fish/info.txt']
+    assert_equal '1234', crate.payload['fish/info.txt'].source.read.chomp
+    assert crate.payload['fish/root.txt']
+    assert crate.payload['fish/data/info.txt']
+    assert crate.payload['fish/data/nested.txt']
+    assert crate.payload['fish/data/binary.jpg']
     assert_equal ['./', 'fish/', 'ro-crate-metadata.jsonld', 'ro-crate-preview.html'], crate.entities.map(&:id).sort
   end
 
@@ -101,22 +101,22 @@ class ReaderTest < Test::Unit::TestCase
     Dir.mktmpdir('test-1234-banana') do |dir|
       crate = ROCrate::Reader.read_zip(fixture_file('directory.zip'), target_dir: dir)
 
-      assert crate.entries['fish/info.txt']
-      assert crate.entries['fish/info.txt'].source.to_s.include?('/test-1234-banana')
+      assert crate.payload['fish/info.txt']
+      assert crate.payload['fish/info.txt'].source.to_s.include?('/test-1234-banana')
     end
   end
 
   test 'reading from directory with directories' do
     crate = ROCrate::Reader.read_directory(fixture_file('directory_crate').path)
 
-    assert crate.entries.values.all? { |e| e.is_a?(ROCrate::Entry) }
-    assert crate.entries['fish/info.txt']
-    assert_equal '1234', crate.entries['fish/info.txt'].source.read.chomp
-    refute crate.entries['fish/root.txt'].directory?
-    assert crate.entries['fish/data'].directory?
-    assert crate.entries['fish/data/info.txt']
-    refute crate.entries['fish/data/nested.txt'].remote?
-    assert crate.entries['fish/data/binary.jpg']
+    assert crate.payload.values.all? { |e| e.is_a?(ROCrate::Entry) }
+    assert crate.payload['fish/info.txt']
+    assert_equal '1234', crate.payload['fish/info.txt'].source.read.chomp
+    refute crate.payload['fish/root.txt'].directory?
+    assert crate.payload['fish/data'].directory?
+    assert crate.payload['fish/data/info.txt']
+    refute crate.payload['fish/data/nested.txt'].remote?
+    assert crate.payload['fish/data/binary.jpg']
     assert_equal ['./', 'fish/', 'ro-crate-metadata.jsonld', 'ro-crate-preview.html'], crate.entities.map(&:id).sort
   end
 
@@ -162,37 +162,38 @@ class ReaderTest < Test::Unit::TestCase
     assert ext_file.source.is_a?(ROCrate::RemoteEntry)
     assert_equal 'http://example.com/external_ref.txt', ext_file.id
     assert_equal 'file contents', ext_file.source.read
+    assert crate.preview.source.source.is_a?(ROCrate::PreviewGenerator)
   end
 
   test 'reading from directory with unlisted files' do
     crate = ROCrate::Reader.read_directory(fixture_file('sparse_directory_crate').path)
 
-    assert_equal 11, crate.entries.count
-    assert crate.entries['listed_file.txt']
-    assert crate.entries['unlisted_file.txt']
-    assert crate.entries['fish']
-    assert_equal '1234', crate.entries['fish/info.txt'].source.read.chomp
-    refute crate.entries['fish/root.txt'].directory?
-    assert crate.entries['fish/data'].directory?
-    assert crate.entries['fish/data/info.txt']
-    refute crate.entries['fish/data/nested.txt'].remote?
-    assert crate.entries['fish/data/binary.jpg']
+    assert_equal 11, crate.payload.count
+    assert crate.payload['listed_file.txt']
+    assert crate.payload['unlisted_file.txt']
+    assert crate.payload['fish']
+    assert_equal '1234', crate.payload['fish/info.txt'].source.read.chomp
+    refute crate.payload['fish/root.txt'].directory?
+    assert crate.payload['fish/data'].directory?
+    assert crate.payload['fish/data/info.txt']
+    refute crate.payload['fish/data/nested.txt'].remote?
+    assert crate.payload['fish/data/binary.jpg']
     assert_equal ['./', 'listed_file.txt', 'ro-crate-metadata.jsonld', 'ro-crate-preview.html'], crate.entities.map(&:id).sort
   end
 
   test 'reading from a zip with unlisted files' do
     crate = ROCrate::Reader.read_zip(fixture_file('sparse_directory_crate.zip').path)
 
-    assert_equal 11, crate.entries.count
-    assert crate.entries['listed_file.txt']
-    assert crate.entries['unlisted_file.txt']
-    assert crate.entries['fish']
-    assert_equal '1234', crate.entries['fish/info.txt'].source.read.chomp
-    refute crate.entries['fish/root.txt'].directory?
-    assert crate.entries['fish/data'].directory?
-    assert crate.entries['fish/data/info.txt']
-    refute crate.entries['fish/data/nested.txt'].remote?
-    assert crate.entries['fish/data/binary.jpg']
+    assert_equal 11, crate.payload.count
+    assert crate.payload['listed_file.txt']
+    assert crate.payload['unlisted_file.txt']
+    assert crate.payload['fish']
+    assert_equal '1234', crate.payload['fish/info.txt'].source.read.chomp
+    refute crate.payload['fish/root.txt'].directory?
+    assert crate.payload['fish/data'].directory?
+    assert crate.payload['fish/data/info.txt']
+    refute crate.payload['fish/data/nested.txt'].remote?
+    assert crate.payload['fish/data/binary.jpg']
     assert_equal ['./', 'listed_file.txt', 'ro-crate-metadata.jsonld', 'ro-crate-preview.html'], crate.entities.map(&:id).sort
   end
 
@@ -201,30 +202,30 @@ class ReaderTest < Test::Unit::TestCase
     string_io.write(::File.read(fixture_file('sparse_directory_crate.zip').path))
     string_io.rewind
     assert string_io.is_a?(StringIO)
-    assert_equal 11, ROCrate::Reader.read_zip(string_io).entries.count
+    assert_equal 11, ROCrate::Reader.read_zip(string_io).payload.count
 
     path = Pathname.new(fixture_file('sparse_directory_crate.zip').path)
     assert path.is_a?(Pathname)
-    assert_equal 11, ROCrate::Reader.read_zip(path).entries.count
+    assert_equal 11, ROCrate::Reader.read_zip(path).payload.count
 
     file = ::File.open(fixture_file('sparse_directory_crate.zip').path)
     assert file.is_a?(::File)
-    assert_equal 11, ROCrate::Reader.read_zip(file).entries.count
+    assert_equal 11, ROCrate::Reader.read_zip(file).payload.count
 
     string = fixture_file('sparse_directory_crate.zip').path
     assert string.is_a?(String)
-    assert_equal 11, ROCrate::Reader.read_zip(string).entries.count
+    assert_equal 11, ROCrate::Reader.read_zip(string).payload.count
   end
 
   test 'reading from zip where the crate root is nested somewhere within' do
     crate = ROCrate::Reader.read_zip(fixture_file('nested_directory.zip'))
 
-    assert crate.entries['fish/info.txt']
-    assert_equal '1234', crate.entries['fish/info.txt'].source.read.chomp
-    assert crate.entries['fish/root.txt']
-    assert crate.entries['fish/data/info.txt']
-    assert crate.entries['fish/data/nested.txt']
-    assert crate.entries['fish/data/binary.jpg']
+    assert crate.payload['fish/info.txt']
+    assert_equal '1234', crate.payload['fish/info.txt'].source.read.chomp
+    assert crate.payload['fish/root.txt']
+    assert crate.payload['fish/data/info.txt']
+    assert crate.payload['fish/data/nested.txt']
+    assert crate.payload['fish/data/binary.jpg']
     assert_equal ['./', 'fish/', 'ro-crate-metadata.json', 'ro-crate-preview.html'], crate.entities.map(&:id).sort
   end
 
@@ -251,4 +252,10 @@ class ReaderTest < Test::Unit::TestCase
                      }
                  ], context
   end
+
+  test 'existing preview is used even if not mentioned in metadata' do
+    crate = ROCrate::Reader.read_zip(fixture_file('biobb_hpc_workflows-condapack.zip').path)
+    assert crate.preview.source.source.is_a?(Pathname)
+    assert_equal 80526, crate.preview.source.read.length
+  end
 end

data/test/test_helper.rb

@@ -6,5 +6,9 @@ require 'ro_crate'
 require 'webmock/test_unit'
 
 def fixture_file(name, *args)
-  ::File.open(::File.join(::File.dirname(__FILE__), 'fixtures', name), *args)
+  ::File.open(::File.join(fixture_dir, name), *args)
+end
+
+def fixture_dir
+  ::File.join(::File.dirname(__FILE__), 'fixtures')
 end

data/test/writer_test.rb

@@ -47,8 +47,12 @@ class WriterTest < Test::Unit::TestCase
   end
 
   test 'writing to zip' do
+    # Remote entries should not be written, so this 500 error should not affect anything.
+    stub_request(:get, 'http://example.com/external_ref.txt').to_return(status: 500)
+
     crate = ROCrate::Crate.new
     crate.add_file(fixture_file('info.txt'))
+    crate.add_file('http://example.com/external_ref.txt')
     crate.add_file(fixture_file('data.csv'), 'directory/data.csv')
 
     Tempfile.create do |file|
@@ -57,6 +61,7 @@ class WriterTest < Test::Unit::TestCase
       Zip::File.open(file) do |zipfile|
         assert zipfile.file.exist?(ROCrate::Metadata::IDENTIFIER)
         assert zipfile.file.exist?(ROCrate::Preview::IDENTIFIER)
+        refute zipfile.file.exist?('external_ref.txt')
         assert_equal 6, zipfile.file.size('info.txt')
         assert_equal 20, zipfile.file.size('directory/data.csv')
       end
@@ -152,4 +157,36 @@ class WriterTest < Test::Unit::TestCase
       end
     end
   end
+
+  test 'writing with conflicting paths in payload obeys specificity rules' do
+    crate = ROCrate::Crate.new
+
+    # Payload from crate
+    crate.add_all(fixture_file('directory').path, false)
+    Dir.mktmpdir do |dir|
+      ROCrate::Writer.new(crate).write(dir)
+
+      assert_equal "5678\n", ::File.read(::File.join(dir, 'data', 'info.txt'))
+    end
+
+    # Payload from crate + directory
+    crate.add_directory(fixture_file('conflicting_data_directory').path.to_s, 'data')
+    Dir.mktmpdir do |dir|
+      ROCrate::Writer.new(crate).write(dir)
+
+      assert_equal 'abcd', ::File.read(::File.join(dir, 'data', 'info.txt')), 'Directory payload should take priority over Crate.'
+      assert_equal "No, I am nested!\n", ::File.read(::File.join(dir, 'data', 'nested.txt')), 'Directory payload should take priority over Crate.'
+      assert ::File.exist?(::File.join(dir, 'data', 'binary.jpg'))
+    end
+
+    # Payload from crate + directory + file
+    crate.add_file(StringIO.new('xyz'), 'data/info.txt')
+    Dir.mktmpdir do |dir|
+      ROCrate::Writer.new(crate).write(dir)
+
+      assert_equal 'xyz', ::File.read(::File.join(dir, 'data', 'info.txt')), 'File payload should take priority over Crate and Directory.'
+      assert_equal "No, I am nested!\n", ::File.read(::File.join(dir, 'data', 'nested.txt')), 'Directory payload should take priority over Crate.'
+      assert ::File.exist?(::File.join(dir, 'data', 'binary.jpg'))
+    end
+  end
 end

metadata

@@ -1,29 +1,35 @@
 --- !ruby/object:Gem::Specification
 name: ro-crate
 version: !ruby/object:Gem::Version
-  version: 0.4.11
+  version: 0.4.12
 platform: ruby
 authors:
 - Finn Bacall
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2021-04-30 00:00:00.000000000 Z
+date: 2021-09-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: addressable
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '2.7'
+    - - "<"
       - !ruby/object:Gem::Version
-        version: 2.7.0
+        version: '2.9'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '2.7'
+    - - "<"
       - !ruby/object:Gem::Version
-        version: 2.7.0
+        version: '2.9'
 - !ruby/object:Gem::Dependency
   name: rubyzip
   requirement: !ruby/object:Gem::Requirement
@@ -146,6 +152,9 @@ files:
 - test/crate_test.rb
 - test/directory_test.rb
 - test/entity_test.rb
+- test/fixtures/biobb_hpc_workflows-condapack.zip
+- test/fixtures/conflicting_data_directory/info.txt
+- test/fixtures/conflicting_data_directory/nested.txt
 - test/fixtures/crate-spec1.1/file with spaces.txt
 - test/fixtures/crate-spec1.1/ro-crate-metadata.json
 - test/fixtures/data.csv