pubid 2.0.0.pre.alpha.13 → 2.0.0.pre.alpha.15

data/lib/pubid/itu/urn_generator.rb

@@ -16,7 +16,10 @@ module Pubid
       def generate_base_urn
         parts = ["urn", "itu"]
 
-        if identifier.sector
+        # An Operational Bulletin is cross-bureau: its sector is a spelling,
+        # not identity (see SpecialPublication#==), so it stays out of the URN.
+        if identifier.sector &&
+            !identifier.is_a?(Identifiers::SpecialPublication)
           sector = identifier.sector.to_s
           parts << sector.to_s.downcase
         else
@@ -56,7 +59,11 @@ module Pubid
 
         if identifier.date
           date = identifier.date
-          if date&.year && date.month
+          # Only an Operational Bulletin's printed date carries a day
+          # ("15.III.2016"); it is in `==`, so it must reach the URN too.
+          if date&.year && date.month && date.day
+            parts << "#{date.day}/#{date.month}/#{date.year}"
+          elsif date&.year && date.month
             parts << "#{date.month}/#{date.year}"
           elsif date&.year
             parts << date.year.to_s

data/lib/pubid/jcgm.rb

@@ -35,7 +35,7 @@ module Pubid
       parser = Parser.new
       builder = Builder.new
 
-      parsed = parser.parse(identifier)
+      parsed = Pubid::Parg::Backend.parse(:jcgm, identifier)
       builder.build(parsed)
     end
 

data/lib/pubid/jis/identifier.rb

@@ -174,7 +174,7 @@ module Pubid
           raise Pubid::Errors::InvalidInputError, Pubid::INPUT_TOO_LONG_MESSAGE
         end
 
-        parsed = Parser.parse(identifier)
+        parsed = Pubid::Parg::Backend.parse(:jis, identifier)
         Builder.build(parsed)
       end
     end

data/lib/pubid/nist/builder.rb

@@ -87,6 +87,7 @@ module Pubid
         # Note: :base_portion is lost during parser merge, so check for supplement indicators
         if parsed_hash[:supplement_date_range] || parsed_hash[:supplement_slash_year] ||
             parsed_hash[:supplement_month_year] || parsed_hash[:supplement_year] ||
+            parsed_hash[:supplement_empty] ||
             parsed_hash[:supplement] || parsed_hash[:base_portion]
           return build_circular_supplement(parsed_hash)
         end

data/lib/pubid/nist/identifiers/base.rb

@@ -696,14 +696,25 @@ module Pubid
           result += "#{vol_str}n#{issue_number.number}"
         end
 
-        # Use edition component - NO space before edition in MR format (per NIST spec)
-        result += edition.to_s if edition
+        # With a number, the edition glues to it per the NIST spec
+        # ("800-53r5"); series-only editions take a dot separator
+        # ("NBS.CIRC.e2" — the attested raw spelling; testsuite#5 C4/C5).
+        result += if edition
+                    number ? edition.to_s : ".#{edition}"
+                  else
+                    ""
+                  end
 
         # Use version_component
         result += version_component.to_s(:mr) if version_component
 
-        # Supplement (e.g. ".9981sup7") - keep distinct documents distinct
-        result += supplement_short
+        # Supplement (e.g. ".9981sup7") - keep distinct documents distinct;
+        # a series-only supplement ("NBS.CIRC.sup") takes the dot itself.
+        result += if supplement
+                    number ? supplement_short : ".#{supplement_short}"
+                  else
+                    ""
+                  end
 
         # Use update_component
         result += update_component.to_s(:mr) if update_component

data/lib/pubid/nist/parser.rb

@@ -726,6 +726,15 @@ module Pubid
             # 4-digit years so it can't swallow "sup3/1926" or a base number.
             ((str("supp") | str("sup")) >> match("[0-9]").repeat(4, 4).as(:supp_year_start) >>
              dash >> match("[0-9]").repeat(4, 4).as(:supp_year_end)).as(:supplement_date_range) |
+            # Bare supplement marker to the whole series, no base number
+            # ("NBS.CIRC.sup" — testsuite#5 C4): the marker alone is the
+            # supplement.
+            ((str("supp") | str("sup")) >>
+              (
+                (month_abbrev >> digits).as(:supplement_month_year) |
+                (digits.as(:supp_number) >> slash >> digits.as(:supp_year)).as(:supplement_slash_year) |
+                str("").as(:supplement_empty)
+              ).maybe) |
             # With base identifier + supplement
             (
               # Capture base portion (everything before "supp" or "sup" or slash+year)

data/lib/pubid/nist/urn_parser.rb

@@ -42,13 +42,22 @@ module Pubid
         type_token, payload = parts
         code, revision = parse_payload(payload)
 
-        text = "NIST #{type_label(type_token)} #{code}"
+        text = "#{publisher_for(type_token)} #{type_label(type_token)} #{code}"
         text += revision if revision
         flavor_parse(text)
       end
 
       private
 
+      # The NBS-era series carry the NBS imprint in their canonical form
+      # ("NBS CSM 1") — the rebuild must name the publisher the document
+      # carries, or the flavor parse rejects its own URN's rebuild.
+      NBS_SERIES = ["csm"].freeze
+
+      def publisher_for(type_token)
+        NBS_SERIES.include?(type_token.downcase) ? "NBS" : "NIST"
+      end
+
       def parse_payload(payload)
         # Strip the trailing ".supp" supplement marker.
         stripped = payload.sub(/\.supp\z/, "")

data/lib/pubid/oasis/identifier.rb

@@ -146,7 +146,7 @@ module Pubid
           raise Pubid::Errors::InvalidInputError, Pubid::INPUT_TOO_LONG_MESSAGE
         end
 
-        parsed = Parser.parse(identifier)
+        parsed = Pubid::Parg::Backend.parse(:oasis, identifier)
         Builder.build(parsed)
       end
 

data/lib/pubid/ogc/identifier.rb

@@ -85,7 +85,7 @@ module Pubid
           raise Pubid::Errors::InvalidInputError, Pubid::INPUT_TOO_LONG_MESSAGE
         end
 
-        Builder.build(Parser.parse(identifier))
+        Builder.build(Pubid::Parg::Backend.parse(:ogc, identifier))
       end
     end
   end

data/lib/pubid/oiml.rb

@@ -37,7 +37,7 @@ module Pubid
       parser = Parser.new
       builder = Builder.new
 
-      parsed = parser.parse(identifier)
+      parsed = Pubid::Parg::Backend.parse(:oiml, identifier)
       builder.build(parsed)
     end
 

data/lib/pubid/omg/identifier.rb

@@ -53,7 +53,7 @@ module Pubid
           raise Pubid::Errors::InvalidInputError, Pubid::INPUT_TOO_LONG_MESSAGE
         end
 
-        parsed = Parser.parse(identifier)
+        parsed = Pubid::Parg::Backend.parse(:omg, identifier)
         Builder.build(parsed)
       end
     end

data/lib/pubid/parg/artifact.rb

@@ -0,0 +1,46 @@
+# frozen_string_literal: true
+
+require "parsanol"
+
+module Pubid
+  module Parg
+    # A baked, checksum-verified PG artifact for one flavor. The artifact
+    # is the parser of record for the flavor: its grammar, tests, entries,
+    # and embedded tables travel together and are verified on load.
+    class Artifact
+      DATA_DIR = File.expand_path("../../../data/parg", __dir__)
+      # The deferred entity atoms resolve from_table references at parse
+      # time from this dir, so the table YAMLs vendor alongside the
+      # baked artifacts.
+      TABLES_DIR = File.join(DATA_DIR, "tables")
+
+      @artifacts = {}
+
+      class << self
+        # Load and memoize the artifact for a flavor (e.g. :iso).
+        def for(flavor)
+          @artifacts[flavor] ||=
+            begin
+              path = File.join(DATA_DIR, "#{flavor}.json")
+              new(Parsanol::PARG::Artifact.load(path, tables_dir: TABLES_DIR))
+            rescue Parsanol::PARG::Error => e
+              raise Pubid::Errors::ParseError,
+                    "PG artifact #{path} rejected: #{e.message}"
+            end
+        end
+      end
+
+      def initialize(artifact)
+        @artifact = artifact
+      end
+
+      def parse(entry, input)
+        @artifact.parse(entry, input)
+      end
+
+      def checksum
+        @artifact.envelope["checksum"]
+      end
+    end
+  end
+end

data/lib/pubid/parg/backend.rb

@@ -0,0 +1,92 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Parg
+    # The PG-artifact parse backend: runs the flavor's baked artifact on
+    # an input and returns the builder-ready attribute hash the flavor's
+    # Builder expects — the same hash its parslet parser produced.
+    module Backend
+      LEAF_KEYS = %i[value line column offset length].freeze
+
+      module_function
+
+      ALL_PARTS_SUFFIX = "(all parts)"
+
+      def parse(flavor, input, entry: "identifier", merge_top_sequence: true)
+        # The parslet grammar base strips the "(all parts)" suffix before
+        # parsing and marks the tree (Grammar#parse / #mark_all_parts);
+        # the artifact backend carries the same contract so every flavor
+        # whose grammar does not itself consume the suffix keeps working.
+        if input.end_with?(ALL_PARTS_SUFFIX)
+          base = input.sub(/\s*\(all parts\)\s*\z/, "")
+          tree = to_builder_hash(Artifact.for(flavor).parse(entry, base), merge: merge_top_sequence)
+          return mark_all_parts(tree)
+        end
+
+        to_builder_hash(Artifact.for(flavor).parse(entry, input), merge: merge_top_sequence)
+      rescue Parsanol::ParseFailed => e
+        raise Pubid::Errors::ParseError.new(e.message, nil,
+                                            input: input,
+                                            flavor: registered_name(flavor))
+      end
+
+      # The registered flavor name (Registry-resolvable; "3gpp" for
+      # Pubid::Tgpp), falling back to the internal symbol when the
+      # flavor is not registered.
+      def registered_name(flavor)
+        mod = Pubid.const_get(flavor.to_s.split("_").map(&:capitalize).join)
+        names = Pubid::Registry.flavor_names.select { |name| Pubid::Registry.get(name) == mod }
+        # A module may register under several names; the longest is the
+        # canonical one ("cen_cenelec", not the "cen" alias).
+        names.max_by(&:length)
+      rescue NameError
+        nil
+      end || flavor.to_s
+
+      def mark_all_parts(tree)
+        case tree
+        when Hash then tree.merge(all_parts: true)
+        when Array then tree.map { |t| t.merge(all_parts: true) }
+        else tree
+        end
+      end
+
+      # The artifact emits the parsanol-tree wire shape: capture leaves
+      # ({value, line, column, offset, length}) carry their text, and the
+      # top-level sequence is a list of capture hashes. The builder-ready
+      # form scalarizes leaves and folds that top-level list into one
+      # attribute hash (last key wins) — parslet's sequence fold.
+      def to_builder_hash(shape, merge: true)
+        normalized = normalize(shape)
+        return normalized.reduce(:merge) if merge && mergeable_sequence?(normalized)
+
+        normalized
+      end
+
+      def normalize(node)
+        case node
+        when Parsanol::Slice then node.content
+        when Hash then normalize_hash(node)
+        when Array then node.map { |item| normalize(item) }
+        else node
+        end
+      end
+
+      def normalize_hash(node)
+        return node[:value] if wire_leaf?(node)
+
+        node.transform_values { |value| normalize(value) }
+      end
+
+      def mergeable_sequence?(value)
+        value.is_a?(Array) && value.all?(Hash)
+      end
+
+      def wire_leaf?(node)
+        return false unless node.size == LEAF_KEYS.size
+
+        LEAF_KEYS.all? { |key| node.key?(key) }
+      end
+    end
+  end
+end

data/lib/pubid/parg.rb

@@ -0,0 +1,8 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Parg
+    autoload :Artifact, "pubid/parg/artifact"
+    autoload :Backend, "pubid/parg/backend"
+  end
+end

data/lib/pubid/pg.rb

@@ -0,0 +1,8 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Pg
+    autoload :Artifact, "pubid/pg/artifact"
+    autoload :Backend, "pubid/pg/backend"
+  end
+end

data/lib/pubid/plateau.rb

@@ -31,8 +31,7 @@ module Pubid
 
       # Apply legacy update_codes normalization first
       normalized = Core::UpdateCodes.apply(input, :plateau)
-      parser = Parser.new
-      parsed = parser.parse(normalized)
+      parsed = Pubid::Parg::Backend.parse(:plateau, normalized)
       Builder.build(parsed)
     end
 

data/lib/pubid/sae/identifiers/base.rb

@@ -16,7 +16,7 @@ module Pubid
           raise Pubid::Errors::InvalidInputError, Pubid::INPUT_TOO_LONG_MESSAGE
         end
 
-        parsed = Parser.parse(input)
+        parsed = Pubid::Parg::Backend.parse(:sae, input)
         Builder.build(parsed)
       end
 

data/lib/pubid/tgpp/identifier.rb

@@ -113,7 +113,7 @@ module Pubid
           raise Pubid::Errors::InvalidInputError, Pubid::INPUT_TOO_LONG_MESSAGE
         end
 
-        parsed = Parser.parse(identifier)
+        parsed = Pubid::Parg::Backend.parse(:tgpp, identifier)
         Builder.build(parsed)
       end
     end

data/lib/pubid/un/identifier.rb

@@ -32,7 +32,7 @@ module Pubid
           raise Pubid::Errors::InvalidInputError, Pubid::INPUT_TOO_LONG_MESSAGE
         end
 
-        parsed = Parser.parse(identifier)
+        parsed = Pubid::Parg::Backend.parse(:un, identifier, merge_top_sequence: false)
         Builder.build(parsed)
       end
     end

data/lib/pubid/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Pubid
-  VERSION = "2.0.0.pre.alpha.13"
+  VERSION = "2.0.0.pre.alpha.15"
 end

data/lib/pubid/w3c/identifier.rb

@@ -122,7 +122,7 @@ module Pubid
           raise Pubid::Errors::InvalidInputError, Pubid::INPUT_TOO_LONG_MESSAGE
         end
 
-        parsed = Parser.parse(identifier)
+        parsed = Pubid::Parg::Backend.parse(:w3c, identifier)
         Builder.build(parsed)
       end
     end

data/lib/pubid/xsf/identifier.rb

@@ -58,7 +58,7 @@ module Pubid
           raise Pubid::Errors::InvalidInputError, Pubid::INPUT_TOO_LONG_MESSAGE
         end
 
-        parsed = Parser.parse(identifier)
+        parsed = Pubid::Parg::Backend.parse(:xsf, identifier)
         Builder.build(parsed)
       end
     end

data/lib/pubid.rb

@@ -17,6 +17,7 @@ require "pubid/prefixes_support"
 module Pubid
   autoload :Schema, "pubid/schema"
   autoload :Conformance, "pubid/conformance"
+  autoload :Parg, "pubid/parg"
   # Upper bound on the length of an identifier string accepted by any +parse+
   # entry point. Real-world standards identifiers are well under 200 characters;
   # this limit exists purely to keep pathological, attacker-controlled inputs

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: pubid
 version: !ruby/object:Gem::Version
-  version: 2.0.0.pre.alpha.13
+  version: 2.0.0.pre.alpha.15
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-09-24 00:00:00.000000000 Z
+date: 2026-10-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: lutaml-model
@@ -38,6 +38,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: parsanol
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.3.57
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.3.57
 description: PubID is a Ruby library for parsing and generating international publication
   identifiers (PubIDs). It provides a simple and flexible API for working with PubIDs,
   allowing developers to easily create, manipulate, and validate PubIDs in their applications.
@@ -60,6 +74,68 @@ files:
 - data/nist/series.yaml
 - data/nist/stages.yaml
 - data/nist/update_codes.yaml
+- data/parg/adobe.json
+- data/parg/aiee.parg
+- data/parg/amca.json
+- data/parg/ansi.json
+- data/parg/api.json
+- data/parg/ashrae.json
+- data/parg/asme.json
+- data/parg/astm.json
+- data/parg/bipm.json
+- data/parg/bsi.json
+- data/parg/calconnect.json
+- data/parg/ccsds.json
+- data/parg/cen_cenelec.json
+- data/parg/cen_cenelec.parg
+- data/parg/cie.json
+- data/parg/csa.json
+- data/parg/doi.json
+- data/parg/easc.json
+- data/parg/ecma.json
+- data/parg/etsi.json
+- data/parg/evs.json
+- data/parg/gb.json
+- data/parg/gost.json
+- data/parg/iala.json
+- data/parg/iana.json
+- data/parg/idf.parg
+- data/parg/iec.json
+- data/parg/ieee.json
+- data/parg/ietf.json
+- data/parg/ire.parg
+- data/parg/isbn.json
+- data/parg/iso.json
+- data/parg/itu.json
+- data/parg/jcgm.json
+- data/parg/jis.json
+- data/parg/nesc.parg
+- data/parg/nist.json
+- data/parg/oasis.json
+- data/parg/ogc.json
+- data/parg/oiml.json
+- data/parg/omg.json
+- data/parg/plateau.json
+- data/parg/sae.json
+- data/parg/tables/bipm_groups.yaml
+- data/parg/tables/bipm_type_codes.yaml
+- data/parg/tables/bipm_type_names_en.yaml
+- data/parg/tables/bipm_type_names_fr.yaml
+- data/parg/tables/directives_supplements_typed_stages.yaml
+- data/parg/tables/directives_typed_stages.yaml
+- data/parg/tables/idf_typed_stages.yaml
+- data/parg/tables/idf_typed_stages_supplements.yaml
+- data/parg/tables/iec_typed_stages.yaml
+- data/parg/tables/iso_publishers.yaml
+- data/parg/tables/organizations.yaml
+- data/parg/tables/tc_types.yaml
+- data/parg/tables/typed_stages.yaml
+- data/parg/tables/typed_stages_supplements.yaml
+- data/parg/tables/wg_types.yaml
+- data/parg/tgpp.json
+- data/parg/un.json
+- data/parg/w3c.json
+- data/parg/xsf.json
 - data/plateau/update_codes.yaml
 - lib/pubid.rb
 - lib/pubid/adobe.rb
@@ -686,6 +762,7 @@ files:
 - lib/pubid/itu/identifiers/errata.rb
 - lib/pubid/itu/identifiers/handbook.rb
 - lib/pubid/itu/identifiers/question.rb
+- lib/pubid/itu/identifiers/radio_regulations.rb
 - lib/pubid/itu/identifiers/recommendation.rb
 - lib/pubid/itu/identifiers/report.rb
 - lib/pubid/itu/identifiers/special_publication.rb
@@ -843,12 +920,16 @@ files:
 - lib/pubid/omg/identifiers/specification.rb
 - lib/pubid/omg/parser.rb
 - lib/pubid/omg/renderer.rb
+- lib/pubid/parg.rb
+- lib/pubid/parg/artifact.rb
+- lib/pubid/parg/backend.rb
 - lib/pubid/parser.rb
 - lib/pubid/parser/common_parse_methods.rb
 - lib/pubid/parser/common_parse_rules.rb
 - lib/pubid/parser/grammar.rb
 - lib/pubid/parsers/base.rb
 - lib/pubid/parsers/mr_string.rb
+- lib/pubid/pg.rb
 - lib/pubid/plateau.rb
 - lib/pubid/plateau/builder.rb
 - lib/pubid/plateau/identifier.rb