glossarist 2.8.17 → 2.9.0

data/lib/glossarist/tasks/sync_model.rb

@@ -0,0 +1,92 @@
+# frozen_string_literal: true
+
+require "json"
+require "fileutils"
+require "net/http"
+require "uri"
+
+module Glossarist
+  module Tasks
+    # Syncs vendored concept-model data artifacts from glossarist/concept-model.
+    #
+    # concept-model is a data-only repo (TTL, JSON-LD, YAML schemas). It is
+    # not a gem. We vendor the small set of artifacts we need (shapes,
+    # context, ontology) and load them at runtime via ShaclValidator.
+    module SyncModel
+      REPO = "glossarist/concept-model"
+      OUT_DIR = File.expand_path("data/concept-model", File.join(__dir__, "..", "..", ".."))
+
+      TARGETS = {
+        "glossarist.context.jsonld" => %w[
+          ontologies/glossarist.context.jsonld
+          glossarist.context.jsonld
+        ].freeze,
+        "glossarist.ttl" => %w[
+          ontologies/glossarist.ttl
+          glossarist.ttl
+        ].freeze,
+        "shapes/glossarist.shacl.ttl" => %w[
+          ontologies/shapes/glossarist.shacl.ttl
+          shapes/glossarist.shacl.ttl
+        ].freeze,
+      }.freeze
+
+      class << self
+        def call(ref: nil)
+          ref ||= latest_tag
+          FileUtils.mkdir_p(File.join(OUT_DIR, "shapes"))
+
+          TARGETS.each do |out_rel, candidates|
+            content = fetch_any(ref, candidates)
+            out_path = File.join(OUT_DIR, out_rel)
+            FileUtils.mkdir_p(File.dirname(out_path))
+            File.write(out_path, content)
+            puts "  ✓ #{out_rel} (#{content.length} bytes)"
+          end
+
+          write_source_manifest(ref)
+          puts "\nSynced #{TARGETS.length} file(s) from #{REPO}@#{ref}."
+        end
+
+        private
+
+        def latest_tag
+          url = URI("https://api.github.com/repos/#{REPO}/releases/latest")
+          req = Net::HTTP::Get.new(url)
+          req["Accept"] = "application/vnd.github+json"
+          JSON.parse(Net::HTTP.start(url.hostname, url.port, use_ssl: true) { |http| http.request(req) }.body)
+            .fetch("tag_name")
+        rescue StandardError => e
+          warn "Could not determine latest concept-model tag: #{e.message}"
+          exit 1
+        end
+
+        def fetch_any(ref, candidates)
+          candidates.each do |path|
+            begin
+              return fetch_file(ref, path)
+            rescue StandardError
+              next
+            end
+          end
+          raise "Could not fetch any of: #{candidates.join(', ')}"
+        end
+
+        def fetch_file(ref, path)
+          url = URI("https://raw.githubusercontent.com/#{REPO}/#{ref}/#{path}")
+          Net::HTTP.get(url)
+        end
+
+        def write_source_manifest(ref)
+          manifest = {
+            "repo" => REPO,
+            "ref" => ref,
+            "syncedAt" => Time.now.utc.iso8601,
+          }
+          File.write(File.join(OUT_DIR, "SOURCE.json"),
+                     JSON.pretty_generate(manifest) + "\n")
+        end
+      end
+    end
+  end
+end

data/lib/glossarist/transforms/concept_to_gloss_transform.rb

@@ -122,10 +122,11 @@ module Glossarist
 
         dd_attrs = if data
                      data.class.detailed_definition_fields.to_h do |field|
-                       [field, build_gloss_definitions(data.public_send(field))]
+                       key = field == :definition ? :definitions : field
+                       [key, build_gloss_definitions(data.public_send(field))]
                      end
                    else
-                     { definition: [], notes: [], examples: [] }
+                     { definitions: [], notes: [], examples: [] }
                    end
 
         sources = build_gloss_sources(data&.sources)

data/lib/glossarist/utilities/uuid.rb

@@ -65,7 +65,11 @@ module Glossarist
         else
           match_data = namespace.match(/\A(\h{8})-(\h{4})-(\h{4})-(\h{4})-(\h{4})(\h{8})\z/)
 
-          unless match_data.present?
+          # `match_data` is a MatchData on success, nil on failure. Do not
+          # use ActiveSupport's `#present?` here — ActiveSupport is not a
+          # declared dependency of this gem, and `#present?` raises
+          # NoMethodError in bundles where it isn't loaded.
+          unless match_data
             raise ArgumentError,
                   "Only UUIDs are valid namespace identifiers"
           end

data/lib/glossarist/v3/concept_data.rb

@@ -6,6 +6,7 @@ module Glossarist
       attribute :sources, V3::ConceptSource,
                 collection: Collections::ConceptSourceCollection,
                 initialize_empty: true
+      attribute :dates, V3::ConceptDate, collection: true
       attribute :definition, V3::DetailedDefinition,
                 collection: Collections::DetailedDefinitionCollection,
                 initialize_empty: true

data/lib/glossarist/v3/concept_date.rb

@@ -0,0 +1,37 @@
+# frozen_string_literal: true
+
+module Glossarist
+  module V3
+    # V3 variant of ConceptDate. The base {Glossarist::ConceptDate} types
+    # `date` as `:date_time` (ISO 8601 datetime only), but the v3 schema
+    # (`concept-model/schemas/v3/concept.yaml`) declares
+    #
+    #   concept_date:
+    #     properties:
+    #       date:
+    #         type: string
+    #         format: date
+    #
+    # which accepts any date string — full ISO 8601 datetime ("2023-01-01T00:00:00+00:00"),
+    # calendar date ("2023-01-01"), or year-only ("2023"). Datasets such as
+    # the IALA Dictionary (datasets/iala-*/concepts/*.yaml) use year-only
+    # strings for accepted/retired lifecycle markers, so the v3 model needs
+    # to round-trip those without losing the value.
+    #
+    # See BUG_REPORT.md for the full investigation.
+    class ConceptDate < Glossarist::ConceptDate
+      attribute :date, :string
+
+      key_value do
+        map :date, to: :date
+        map :type, to: :type
+      end
+
+      # The `date` is already a free-form string in V3, so the wire form
+      # is the value itself — no iso8601 coercion.
+      def to_yaml_date
+        date
+      end
+    end
+  end
+end

data/lib/glossarist/v3/localized_concept.rb

@@ -4,6 +4,15 @@ module Glossarist
   module V3
     class LocalizedConcept < Glossarist::LocalizedConcept
       attribute :data, V3::ConceptData, default: -> { V3::ConceptData.new }
+
+      def date_accepted_from_yaml(model, value)
+        return if model.date_accepted
+
+        model.data.dates ||= []
+        model.data.dates << V3::ConceptDate.of_yaml(
+          { "date" => value, "type" => "accepted" },
+        )
+      end
     end
   end
 end

data/lib/glossarist/v3/managed_concept.rb

@@ -5,6 +5,8 @@ module Glossarist
     class ManagedConcept < Glossarist::ManagedConcept
       attribute :data, V3::ManagedConceptData, default: -> { V3::ManagedConceptData.new }
       attribute :related, V3::RelatedConcept, collection: true
+      attribute :dates, V3::ConceptDate, collection: true
+      attribute :date_accepted, V3::ConceptDate
       attribute :sources, V3::ConceptSource, collection: true
 
       key_value do
@@ -19,6 +21,12 @@ module Glossarist
         map :schema_version, to: :schema_version
         map :sources, to: :sources
       end
+
+      def date_accepted_from_yaml(model, value)
+        model.date_accepted = V3::ConceptDate.of_yaml(
+          { "date" => value, "type" => "accepted" },
+        )
+      end
     end
   end
 end

data/lib/glossarist/v3.rb

@@ -4,6 +4,7 @@ module Glossarist
   module V3
     autoload :Configuration, "glossarist/v3/configuration"
     autoload :Citation, "glossarist/v3/citation"
+    autoload :ConceptDate, "glossarist/v3/concept_date"
     autoload :ConceptSource, "glossarist/v3/concept_source"
     autoload :DetailedDefinition, "glossarist/v3/detailed_definition"
     autoload :ConceptRef, "glossarist/v3/concept_ref"
@@ -15,6 +16,7 @@ module Glossarist
     autoload :ConceptDocument, "glossarist/v3/concept_document"
 
     Configuration.register_model(Citation, id: :citation)
+    Configuration.register_model(ConceptDate, id: :concept_date)
     Configuration.register_model(ConceptSource, id: :concept_source)
     Configuration.register_model(DetailedDefinition, id: :detailed_definition)
     Configuration.register_model(ConceptData, id: :concept_data)

data/lib/glossarist/validation/rules/related_concept_cycle_rule.rb

@@ -54,11 +54,25 @@ module Glossarist
           graph
         end
 
+        # Returns a graph node identifier for the target of a relationship.
+        #
+        # For intra-edition references (ref.source is nil), the target is
+        # just ref.id — a clause identifier unique within the dataset.
+        #
+        # Returns nil for cross-edition references (ref.source is a URN,
+        # e.g. a +supersedes+ edge pointing at the previous edition's
+        # concept with the same clause id). Such edges cannot form
+        # cycles within the current dataset and are excluded from the
+        # graph. Without this exclusion, a cross-edition edge from
+        # concept 3.1.1.1 to its predecessor in another edition (which
+        # also has identifier 3.1.1.1) would look like a self-loop and
+        # trip the cycle detector with a false positive.
         def resolve_target_id(rel)
           ref = rel.ref
           return nil unless ref
+          return nil if ref.source
 
-          ref.id || ref.source
+          ref.id
         end
 
         def detect_cycles(graph)

data/lib/glossarist/validation/shacl_validator.rb

@@ -0,0 +1,97 @@
+# frozen_string_literal: true
+
+require "rdf/turtle"
+require "shacl"
+require "pathname"
+
+module Glossarist
+  module Validation
+    # Validates Turtle output against SHACL shapes.
+    #
+    # Shapes resolution order:
+    #   1. Explicit :shapes_path option
+    #   2. Vendored shapes at data/concept-model/shapes/glossarist.shacl.ttl
+    #      (shipped with the gem, relative to the gem root)
+    class ShaclValidator
+      VENDORED_SHAPES_PATH =
+        File.expand_path(
+          "../../../data/concept-model/shapes/glossarist.shacl.ttl",
+          __dir__
+        ).freeze
+
+      attr_reader :shapes_path
+
+      def initialize(shapes_path: nil)
+        @shapes_path = shapes_path || self.class.default_shapes_path
+      end
+
+      def validate_files(paths)
+        validate_graphs(Array(paths).map { |p| load_graph(p) })
+      end
+
+      def validate_graphs(graphs)
+        shapes = SHACL::Shapes.from_graph(load_graph(@shapes_path))
+        failures = []
+
+        graphs.each do |graph|
+          report = shapes.execute(graph)
+          next if report.conform?
+
+          failures << Report.new(results: report.results)
+        end
+
+        AggregateReport.new(failures: failures)
+      end
+
+      class << self
+        def default_shapes_path
+          return VENDORED_SHAPES_PATH if File.exist?(VENDORED_SHAPES_PATH)
+
+          raise ArgumentError,
+                "No SHACL shapes path provided and the vendored shapes " \
+                "file (#{VENDORED_SHAPES_PATH}) is missing. Pass " \
+                "shapes_path: '/path/to/glossarist.shacl.ttl'."
+        end
+      end
+
+      private
+
+      def load_graph(path)
+        return path if path.is_a?(RDF::Graph)
+
+        graph = RDF::Graph.new
+        RDF::Turtle::Reader.new(File.read(path)) do |reader|
+          reader.each_statement { |stmt| graph << stmt }
+        end
+        graph
+      end
+    end
+
+    Report = Struct.new(:results, keyword_init: true) do
+      def conformant?
+        results.empty?
+      end
+
+      def to_s
+        "  #{results.length} violation(s)"
+      end
+    end
+
+    AggregateReport = Struct.new(:failures, keyword_init: true) do
+      def conformant?
+        failures.empty?
+      end
+
+      def to_s
+        lines = ["SHACL validation failed: #{failures.length} file(s) with violations"]
+        failures.each do |failure|
+          lines << failure.to_s
+          failure.results.each do |result|
+            lines << "    #{result.path}: #{result.message}"
+          end
+        end
+        lines.join("\n")
+      end
+    end
+  end
+end

data/lib/glossarist/validation.rb

@@ -8,5 +8,6 @@ module Glossarist
     autoload :Rules,             "glossarist/validation/rules"
     autoload :BibliographyIndex, "glossarist/validation/bibliography_index"
     autoload :AssetIndex,        "glossarist/validation/asset_index"
+    autoload :ShaclValidator,    "glossarist/validation/shacl_validator"
   end
 end

data/lib/glossarist/version.rb

@@ -4,5 +4,5 @@
 #
 
 module Glossarist
-  VERSION = "2.8.17"
+  VERSION = "2.9.0"
 end

data/lib/glossarist.rb

@@ -23,6 +23,7 @@ module Glossarist
   autoload :ConceptData,              "glossarist/concept_data"
   autoload :ConceptRef,               "glossarist/concept_ref"
   autoload :ConceptReference,         "glossarist/concept_reference"
+  autoload :Reference, "glossarist/reference"
   autoload :ReferenceExtractor,       "glossarist/reference_extractor"
   autoload :ReferenceResolver,        "glossarist/reference_resolver"
   autoload :ResolutionAdapter,        "glossarist/resolution_adapter"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: glossarist
 version: !ruby/object:Gem::Version
-  version: 2.8.17
+  version: 2.9.0
 platform: ruby
 authors:
 - Ribose
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-06-21 00:00:00.000000000 Z
+date: 2026-07-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: lutaml-model
@@ -52,6 +52,20 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '2.3'
+- !ruby/object:Gem::Dependency
+  name: rdf-turtle
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.3'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.3'
 - !ruby/object:Gem::Dependency
   name: relaton
   requirement: !ruby/object:Gem::Requirement
@@ -92,6 +106,20 @@ dependencies:
     - - "<"
       - !ruby/object:Gem::Version
         version: '3'
+- !ruby/object:Gem::Dependency
+  name: shacl
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.4'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.4'
 - !ruby/object:Gem::Dependency
   name: sts
   requirement: !ruby/object:Gem::Requirement
@@ -170,6 +198,11 @@ files:
 - README.adoc
 - Rakefile
 - config.yml
+- data/concept-model/README.md
+- data/concept-model/SOURCE.json
+- data/concept-model/glossarist.context.jsonld
+- data/concept-model/glossarist.ttl
+- data/concept-model/shapes/glossarist.shacl.ttl
 - exe/glossarist
 - glossarist.gemspec
 - lib/glossarist.rb
@@ -276,6 +309,7 @@ files:
 - lib/glossarist/rdf/gloss_pronunciation.rb
 - lib/glossarist/rdf/gloss_reference.rb
 - lib/glossarist/rdf/localized_literal.rb
+- lib/glossarist/rdf/lutaml_ext.rb
 - lib/glossarist/rdf/namespaces.rb
 - lib/glossarist/rdf/namespaces/dcterms_namespace.rb
 - lib/glossarist/rdf/namespaces/glossarist_namespace.rb
@@ -288,6 +322,7 @@ files:
 - lib/glossarist/rdf/relationship_predicates.rb
 - lib/glossarist/rdf/v3.rb
 - lib/glossarist/rdf/v3/configuration.rb
+- lib/glossarist/reference.rb
 - lib/glossarist/reference_extractor.rb
 - lib/glossarist/reference_resolver.rb
 - lib/glossarist/register_data.rb
@@ -313,6 +348,9 @@ files:
 - lib/glossarist/sts/term_mapper.rb
 - lib/glossarist/table.rb
 - lib/glossarist/table_reference.rb
+- lib/glossarist/tasks/shacl.rake
+- lib/glossarist/tasks/sync.rake
+- lib/glossarist/tasks/sync_model.rb
 - lib/glossarist/transforms.rb
 - lib/glossarist/transforms/concept_to_gloss_transform.rb
 - lib/glossarist/transforms/concept_to_tbx_transform.rb
@@ -339,6 +377,7 @@ files:
 - lib/glossarist/v3.rb
 - lib/glossarist/v3/citation.rb
 - lib/glossarist/v3/concept_data.rb
+- lib/glossarist/v3/concept_date.rb
 - lib/glossarist/v3/concept_document.rb
 - lib/glossarist/v3/concept_ref.rb
 - lib/glossarist/v3/concept_source.rb
@@ -401,6 +440,7 @@ files:
 - lib/glossarist/validation/rules/source_urn_format_rule.rb
 - lib/glossarist/validation/rules/terms_presence_rule.rb
 - lib/glossarist/validation/rules/uuid_format_rule.rb
+- lib/glossarist/validation/shacl_validator.rb
 - lib/glossarist/validation/validation_issue.rb
 - lib/glossarist/validation_result.rb
 - lib/glossarist/version.rb