pubid 2.0.0.pre.alpha.8 → 2.0.0.pre.alpha.11

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

@@ -8,7 +8,7 @@ module Pubid
       # - "NBS CRPL 1-2_3-1" = Reports 1-2 and 3-1 jointly
       # - "NBS CRPL 4-m-5" = Report 4, month m, issue 5
       # - "NBS CRPL c4-4" = Report c4, issue 4
-      class CrplReport < Base
+      class CrplReport < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["CRPL", "NBS CRPL", "CRPL-F-B", "CRPL-F-A", "NBS CRPL-F-B",
@@ -88,6 +88,12 @@ module Pubid
         end
 
         def to_s(format = nil)
+          # A caller passing `annotated:` reaches this positional parameter as
+          # a Hash — Ruby converts keywords for a method that accepts none.
+          # The same idiom `Identifiers::Base#to_s` uses. This renderer
+          # ignores `format`, so only the flag is read.
+          annotated = format.is_a?(Hash) ? format[:annotated] : nil
+
           # Use actual series attribute if it contains subseries (e.g., "CRPL-F-B")
           # Otherwise use default series_code ("CRPL")
           series_to_render = if series&.value&.include?("CRPL-F-")
@@ -124,7 +130,7 @@ module Pubid
           result += append_short_components
 
           result += range_notation if range_notation
-          result
+          annotate_plain_render(result, annotated: annotated)
         end
       end
     end

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

@@ -7,7 +7,7 @@ module Pubid
       # Examples:
       # - "NIST 2022-04-15 001" (short) / "NIST.2022-04-15.001" (mr/doi)
       #   DOI 10.6028/NIST.2022-04-15.001
-      class DatedDocument < Base
+      class DatedDocument < Identifier
         # Empty so the bare "NIST" abbr never shadows a series lookup; the
         # Router guard (parsed_hash[:dated_date] && :dated_seq) is the dispatch path.
         TYPED_STAGES = [].freeze

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

@@ -7,7 +7,7 @@ module Pubid
       # Examples:
       # - "FIPS 140-3" (no NIST prefix)
       # - "NIST FIPS 140-3" (also accepted, normalizes to FIPS 140-3)
-      class FederalInformationProcessingStandards < Base
+      class FederalInformationProcessingStandards < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["FIPS", "NIST FIPS"],

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

@@ -7,7 +7,7 @@ module Pubid
       # Examples:
       # - "NIST GCR 17-917-45" - Basic 3-part number
       # - "NIST GCR 21-917-48v3B" - With volume and letter suffix
-      class GrantContractorReport < Base
+      class GrantContractorReport < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["GCR", "NIST GCR"],

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

@@ -7,7 +7,7 @@ module Pubid
       # Examples:
       # - "NIST HB 130" = Handbook 130
       # - "NBS HB 133" = NBS Handbook 133
-      class Handbook < Base
+      class Handbook < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["HB", "NIST HB", "NBS HB"],
@@ -34,8 +34,15 @@ module Pubid
         # Override to_s to handle edition+year pattern with dash instead of dot
         # Pattern: "44e2-1955" (dash separator) not "44e2.1955" (dot separator)
         def to_s(format = nil)
+          # The rewrite below is `$`-anchored, so it has to run on the PLAIN
+          # string: with the flag set, `super` returns a string ending in
+          # "</span>" and the anchor no longer matches. Render plain, rewrite,
+          # then annotate.
+          annotated = format.is_a?(Hash) ? format[:annotated] : nil
+          format = format[:format] if format.is_a?(Hash)
+
           # Use the base to_s method first
-          result = super
+          result = super(format)
 
           # If edition has additional_text (year), replace dot with dash
           if edition&.additional_text && !edition.additional_text.empty?
@@ -43,7 +50,7 @@ module Pubid
                                  "#{edition.type}#{edition.id}-#{edition.additional_text}")
           end
 
-          result
+          annotate_plain_render(result, annotated: annotated)
         end
       end
     end

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

@@ -8,7 +8,7 @@ module Pubid
       # - "NIST IR 8115" = Interagency Report 8115
       # - "NIST IR 8115r1" = Interagency Report 8115 revision 1
       # - "NBS IR 73-197" = NBS Interagency Report 73-197
-      class InteragencyReport < Base
+      class InteragencyReport < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["IR", "NIST IR", "NBS IR"],
@@ -37,6 +37,11 @@ module Pubid
         # @param format [Symbol, Hash, nil] output format
         # @return [String] formatted identifier
         def to_s(format = nil)
+          # `super` is handed a SYMBOL below, so the base's annotation hook
+          # never sees the flag. Capture it here and wrap what comes back —
+          # exactly once, because the base returned plain text.
+          annotated = format.is_a?(Hash) ? format[:annotated] : nil
+
           # Handle both keyword argument (hash) and positional argument (symbol/string)
           effective_format = if format.is_a?(Hash)
                                format[:format]
@@ -45,11 +50,13 @@ module Pubid
                              end
 
           # If explicit format is specified, use it. Otherwise, default to short.
-          if effective_format.nil?
-            super(:short)
-          else
-            super(effective_format)
-          end
+          rendered = if effective_format.nil?
+                       super(:short)
+                     else
+                       super(effective_format)
+                     end
+
+          annotate_plain_render(rendered, annotated: annotated)
         end
       end
     end

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

@@ -11,7 +11,7 @@ module Pubid
       # - "NBS LC 378sup12/1926" - With supplement date
       # - "NBS LC 378r11/1925" - With revision date
       # - "NBS LC 378(sp)" - With language code
-      class LetterCircular < Base
+      class LetterCircular < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["LCIRC"], # Single definition of truth

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

@@ -8,7 +8,7 @@ module Pubid
       # - "NBS MP 39e1" - Edition with "e" notation
       # - "NBS MP 260e1965" - Edition with year as edition ID
       # NOTE: Parenthetical edition format (e.g., "39(1)") does NOT exist for MP identifiers
-      class MiscellaneousPublication < Base
+      class MiscellaneousPublication < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["MP", "NBS MP"],
@@ -36,13 +36,21 @@ module Pubid
           "MP"
         end
 
+        # A caller passing `annotated:` reaches this positional parameter as a
+        # Hash — Ruby converts keywords for a method that accepts none. The same
+        # idiom `Identifiers::Base#to_s` uses.
         def to_s(format = :short)
-          case format
-          when :mr
-            to_mr_style
-          else
-            to_short_style
-          end
+          annotated = format.is_a?(Hash) ? format[:annotated] : nil
+          format = format[:format] || :short if format.is_a?(Hash)
+
+          rendered = case format
+                     when :mr
+                       to_mr_style
+                     else
+                       to_short_style
+                     end
+
+          annotate_plain_render(rendered, annotated: annotated)
         end
 
         public

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

@@ -10,7 +10,7 @@ module Pubid
       # - "NBS MONO 128pt1" - Monograph with part
       # - "NIST MONO 1-1F" - Monograph with letter suffix
       # - "NIST MONO 1-2Bv1" - Monograph with letter suffix and volume
-      class Monograph < Base
+      class Monograph < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["MONO", "NBS MONO", "NIST MONO"],
@@ -35,6 +35,11 @@ module Pubid
         end
 
         def to_s(format = nil)
+          # `super` is handed a SYMBOL below, so the base's annotation hook
+          # never sees the flag. Capture it here and wrap what comes back —
+          # exactly once, because the base returned plain text.
+          annotated = format.is_a?(Hash) ? format[:annotated] : nil
+
           # Handle both keyword argument (hash) and positional argument (symbol/string)
           effective_format = if format.is_a?(Hash)
                                format[:format]
@@ -43,13 +48,15 @@ module Pubid
                              end
 
           # If explicit format is specified, use it. Otherwise, default to short.
-          if effective_format.nil?
-            super(:short)
-          elsif effective_format == :mr
-            to_mr_style
-          else
-            super(effective_format)
-          end
+          rendered = if effective_format.nil?
+                       super(:short)
+                     elsif effective_format == :mr
+                       to_mr_style
+                     else
+                       super(effective_format)
+                     end
+
+          annotate_plain_render(rendered, annotated: annotated)
         end
 
         private

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

@@ -8,7 +8,7 @@ module Pubid
       # - "NIST NCSTAR 1-1Cv1" → "NIST NCSTAR 1-1Cv1" (round-trips)
       # - "NIST NCSTAR 1-1b" → "NIST NCSTAR 1-1B" (normalizes letter)
       # - "NIST NCSTAR 1-1cv1" → "NIST NCSTAR 1-1Cv1" (normalizes letter, keeps v1 format)
-      class Ncstar < Base
+      class Ncstar < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["NCSTAR", "NIST NCSTAR"],
@@ -34,7 +34,7 @@ module Pubid
 
         # NCSTAR-specific: do NOT expand Cv patterns to ", Volume" format
         # The base class would render "1-1Cv1" as is, which is correct for NCSTAR
-        # No custom rendering needed - just inherit from Base
+        # No custom rendering needed - just inherit from Identifier
       end
     end
   end

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

@@ -8,7 +8,7 @@ module Pubid
       # - "NBS NSRDS 1" - Basic NSRDS
       # - "NSRDS-NBS 1" - With hyphen prefix
       # - "NBS NSRDS 1p1" - With part notation
-      class Nsrds < Base
+      class Nsrds < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["NSRDS", "NBS NSRDS"],

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

@@ -6,7 +6,7 @@ module Pubid
       # NIST OWMWP (Office of Weights and Measures Workshop Proceedings)
       # Examples:
       # - "NIST OWMWP 06-13-2018" - Date-based format MM-DD-YYYY
-      class Owmwp < Base
+      class Owmwp < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["OWMWP", "NIST OWMWP"],

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

@@ -11,7 +11,7 @@ module Pubid
       # - "NBS RPT ADHOC" - Ad hoc report
       # - "NBS RPT div9" - Division report
       # - "NBS RPT 4817-A" - Report with letter suffix
-      class Report < Base
+      class Report < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["RPT", "NBS RPT"],
@@ -39,13 +39,21 @@ module Pubid
           "RPT"
         end
 
+        # A caller passing `annotated:` reaches this positional parameter as a
+        # Hash — Ruby converts keywords for a method that accepts none. The same
+        # idiom `Identifiers::Base#to_s` uses.
         def to_s(format = :short)
-          case format
-          when :mr
-            to_mr_style
-          else
-            to_short_style
-          end
+          annotated = format.is_a?(Hash) ? format[:annotated] : nil
+          format = format[:format] || :short if format.is_a?(Hash)
+
+          rendered = case format
+                     when :mr
+                       to_mr_style
+                     else
+                       to_short_style
+                     end
+
+          annotate_plain_render(rendered, annotated: annotated)
         end
 
         private

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

@@ -8,7 +8,16 @@ module Pubid
       # - "NIST SP 800-53" = Special Publication 800-53
       # - "NIST SP 800-53r5" = Special Publication 800-53 revision 5
       # - "NIST SP 800-57pt1r4" = Special Publication 800-57 part 1 revision 4
-      class SpecialPublication < Base
+      class SpecialPublication < Identifier
+        # Per Appendix A.2 of the NIST PubID Syntax (April 2022), the SP series
+        # is subdivided into named subseries. The subseries number is the
+        # leading token of a report-num like "800-53" or "1190GB-12". The
+        # Builder exposes this as `identifier.subseries` so callers can query
+        # which subseries a document belongs to.
+        SP_SUBSERIES = %w[
+          250 260 300 400 480 500 700 800 823 960 1190GB 1200 1500 1800 1900 2000 2100
+        ].freeze
+
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["SP", "NIST SP", "NBS SP"], # Compound series

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

@@ -7,7 +7,7 @@ module Pubid
       # Examples:
       # - "NIST TN 1234" = Technical Note 1234
       # - "NBS TN 567" = NBS Technical Note 567
-      class TechnicalNote < Base
+      class TechnicalNote < Identifier
         TYPED_STAGES = [
           Pubid::Components::TypedStage.new(
             abbr: ["TN", "NIST TN", "NBS TN"],

data/lib/pubid/nist/identifiers.rb

@@ -3,7 +3,6 @@
 module Pubid
   module Nist
     module Identifiers
-      autoload :Base, "#{__dir__}/identifiers/base"
       autoload :Circular, "#{__dir__}/identifiers/circular"
       autoload :CircularSupplement, "#{__dir__}/identifiers/circular_supplement"
       autoload :CommercialStandard, "#{__dir__}/identifiers/commercial_standard"

data/lib/pubid/nist/parser.rb

@@ -6,7 +6,7 @@ module Pubid
   module Nist
     # Parser class for NIST identifiers
     # Single Responsibility: Parsing NIST identifier syntax
-    class Parser < Parslet::Parser
+    class Parser < ::Pubid::Parser::Grammar
 
       # Class-level parse method with preprocessing.
       # Delegates all string normalization to Nist::Preprocessor, then
@@ -346,6 +346,16 @@ module Pubid
           match("[0-9]").repeat(1).as(:dated_seq)
       end
 
+      # "NIST Research Library (YYYY)" — one-off NIST publication that doesn't
+      # fit the standard series+number shape (no report number, year in parens).
+      # See pubid/pubid#151.
+      rule(:research_library_identifier) do
+        publisher >> space >>
+          str("Research Library").as(:series) >>
+          (space >> str("(") >>
+            match("[0-9]").repeat(4, 4).as(:year) >> str(")")).maybe
+      end
+
       # LEGACY EDITION PATTERNS (for backward compatibility during migration)
       # These will be gradually replaced as we migrate to proper Edition/Date components
       rule(:legacy_edition) do
@@ -438,7 +448,12 @@ module Pubid
 
       # Volume
       rule(:volume) do
-        (space.maybe >> (str("v") | str(" Vol. "))) >>
+        # Long-form "Vol." comes FIRST so PEG ordered choice doesn't commit to
+        # the bare "v" alternative after consuming the leading space via
+        # space.maybe (which would then starve the " Vol. " form). Both
+        # "Vol. 4" (period + space) and "Vol.4" (period only) are accepted.
+        (space.maybe >>
+          (str("Vol. ") | str("Vol.") | str("v"))) >>
           (digits >>
            # Support letter ranges (lowercase normalized in preprocessing)
            (str("a-l") | str("m-z") | str("A-L") | str("M-Z")).maybe >>
@@ -448,7 +463,8 @@ module Pubid
 
       # Part - enhanced to support patterns like p1adde1 AND pt3r1 (part with revision)
       rule(:part) do
-        ((space.maybe >> (str("pt") | str("p") | str("P"))) | str(" Part ")) >>
+        (str(" Part. ") | str(" Part.") | str(" Part ") |
+         (space.maybe >> (str("pt") | str("p") | str("P")))) >>
           (digits >>
            # NEW: Revision after part number: pt3r1, p1r1 (space.maybe for preprocessing)
            (space.maybe >> str("r") >> digits).maybe >>
@@ -474,7 +490,8 @@ module Pubid
           # Enhanced to accept letter-only revisions and space before r
           # ENHANCED: Accept BOTH lowercase and uppercase letters in suffix
           # ENHANCED: Capture original format prefix for format preservation (e.g., " Rev. 5")
-          ((str(" rev ") | str("rev") | str(" r") | str("r") | str(" Rev. ") | str(" Revision (r)")).as(:revision_prefix) >>
+          ((str(" rev ") | str("rev") | str(" r") | str("r") | str(" Rev. ") | str(" Rev.") | str(" Revision (r)")).as(:revision_prefix) >>
+           space.maybe >>
             ((digits >> match("[a-zA-Z]").maybe) | match("[a-zA-Z]").repeat(1)).as(:revision_id)).as(:revision) |
           # NEW: Standalone 'r' - MUST BE LAST to avoid consuming from other patterns
           # Matches " r" at end of input (after preprocessing: "800-56a r", "800-27 r")
@@ -656,6 +673,7 @@ module Pubid
         circ_supplement_identifier |
           dated_identifier |
           mr_identifier |
+          research_library_identifier |
           (
             # Compound series (includes publisher in series name)
             compound_series >> (space | dot) >>
@@ -690,6 +708,11 @@ module Pubid
             # Date range supplement (no base number)
             (str("supp") >> month_abbrev.as(:supp_month_start) >> digits.as(:supp_year_start) >>
              dash >> month_abbrev.as(:supp_month_end) >> digits.as(:supp_year_end)).as(:supplement_date_range) |
+            # Year-only date range supplement to the whole series, no months
+            # (e.g. "NBS CIRC sup1925-1926" — pubid/pubid#152). Requires two
+            # 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) |
             # With base identifier + supplement
             (
               # Capture base portion (everything before "supp" or "sup" or slash+year)

data/lib/pubid/nist/preprocessor.rb

@@ -343,6 +343,21 @@ module Pubid
         revert_handbook_edition!
         revert_owmwp_date!
         revert_report_year_range!
+        revert_circ_supplement_year_range!
+      end
+
+      # CIRC/LCIRC year-only series supplements: "NBS CIRC sup1925-1926" is a
+      # supplement spanning a year range, so the trailing "-1926" is a range
+      # end, not an edition. convert_dashyear_to_edition! would have rewritten
+      # it to "sup1925e1926"; restore the dash so the circ_supplement_identifier
+      # grammar can read the range (pubid/pubid#152).
+      def revert_circ_supplement_year_range!
+        return unless @cleaned.include?("CIRC")
+
+        # Anchor on the series prefix ("CIRC " / "CIRC." / "LCIRC ") so only the
+        # whole-series form (sup right after the series, no base number) is
+        # reverted — a based "24sup1940e1950" keeps its genuine edition.
+        @cleaned = @cleaned.gsub(/(CIRC[ .])(supp?)(\d{4})e(\d{4})(?=\s|$)/, '\1\2\3-\4')
       end
 
       # HB handbooks: "HB 130e1979" → "HB 130-1979" (year is part of

data/lib/pubid/nist/router.rb

@@ -127,10 +127,10 @@ module Pubid
         # instead of an injected dependency.
         typed_stage = Pubid::Nist.locate_stage(series)
         if typed_stage
-          Pubid::Nist.locate_type(typed_stage.type_code) || Identifiers::Base
+          Pubid::Nist.locate_type(typed_stage.type_code) || Identifier
         else
           # Fallback to Base for unmapped series (e.g., "AMS", "VTS")
-          Identifiers::Base
+          Identifier
         end
       end
 

data/lib/pubid/nist/series/ir.rb

@@ -36,8 +36,6 @@ module Pubid
           )
           edition_obj = Components::Edition.new(type: "r", id: "1")
           identifier.edition = edition_obj
-          identifier.edition_component = edition_obj
-          identifier.revision = "r1"
           true
         end
 
@@ -52,7 +50,6 @@ module Pubid
             value: "#{match[1]}-#{match[2]}",
           )
           identifier.edition = nil
-          identifier.edition_component = nil
         end
       end
     end

data/lib/pubid/nist/supplement_identifier.rb

@@ -6,34 +6,49 @@ module Pubid
     # Supplements wrap a base identifier with additional edition information
     #
     # Architecture follows ISO pattern:
-    # - base_identifier: The base document being supplemented
+    # - base: The base document being supplemented
     # - edition: Edition information for the supplement
     #
     # Examples:
     # - "NBS CIRC 101e2supp" → CircularSupplement(base: "NBS CIRC 101", edition: e2)
     # - "NBS CIRC 25supp-1924" → CircularSupplement(base: "NBS CIRC 25", edition: 1924)
-    class SupplementIdentifier < Identifiers::Base
-      attribute :base_identifier, Identifiers::Base, polymorphic: true
+    class SupplementIdentifier < Identifier
+      attribute :base, Identifier, polymorphic: true
 
-      # Delegate publisher to base_identifier
+      # Delegate publisher to base
       def publisher
-        base_identifier&.publisher
+        base&.publisher
       end
 
-      # Delegate series to base_identifier
+      # Delegate series to base
       def series
-        base_identifier&.series
+        base&.series
       end
 
+      # Four exits — a date range, a base-less delegation, an update form and
+      # the canonical one — so the composition stays whole in its own method
+      # and `to_s` wraps its single result. `super` is not reachable from a
+      # private method, so it is handed in as a block.
       def to_s(format = :short)
+        annotated = format.is_a?(Hash) ? format[:annotated] : nil
+        format = format[:format] || :short if format.is_a?(Hash)
+
+        rendered = render_plain(format) { super(format) }
+
+        annotate_plain_render(rendered, annotated: annotated)
+      end
+
+      private
+
+      def render_plain(format)
         # Handle date range supplements (no base identifier)
         if supplement_date_range_start && supplement_date_range_end
           return "NBS CIRC sup#{supplement_date_range_start}-#{supplement_date_range_end}"
         end
 
-        return super unless base_identifier
+        return yield unless base
 
-        result = base_identifier.to_s(format)
+        result = base.to_s(format)
 
         # NEW: Handle update attribute (e.g., "Upd12-1926" for supplement patterns)
         if update

data/lib/pubid/nist.rb

@@ -1,5 +1,6 @@
 # frozen_string_literal: true
 
+require "pubid"
 module Pubid
   module Nist
     extend Pubid::PrefixesSupport
@@ -21,7 +22,7 @@ module Pubid
     autoload :CircularSupplementBuilder, "#{__dir__}/nist/circular_supplement_builder"
     autoload :Components, "#{__dir__}/nist/components"
     autoload :Configuration, "#{__dir__}/nist/configuration"
-    autoload :Identifier, "#{__dir__}/nist/identifier"
+    autoload :Identifier, "#{__dir__}/nist/identifiers/base"
     autoload :Identifiers, "#{__dir__}/nist/identifiers"
     autoload :Parser, "#{__dir__}/nist/parser"
     autoload :ParserOutputNormalizer, "#{__dir__}/nist/parser_output_normalizer"
@@ -37,7 +38,7 @@ module Pubid
     #
     # NIST identifier classes use `typed_stages` (not the `self.type` Hash
     # pattern used by CEN/JCGM/ANSI), so they cannot be auto-discovered.
-    # Identifiers::Base is the fallback for unmapped series (e.g. AMS, VTS)
+    # Identifier is the fallback for unmapped series (e.g. AMS, VTS)
     # and must be excluded from typed-stage aggregation.
     IDENTIFIER_TYPES = [
       Identifiers::SpecialPublication,
@@ -60,15 +61,20 @@ module Pubid
       Identifiers::CommercialStandardEmergency,
       Identifiers::CommercialStandardsMonthly,
       Identifiers::DatedDocument,
-      Identifiers::Base, # Fallback for unmapped series
+      Identifier, # Fallback for unmapped series
     ].freeze
 
     # Parse a NIST identifier string
     # @param identifier [String] the identifier string to parse
-    # @return [Identifiers::Base] the parsed identifier
+    # @return [Identifier] the parsed identifier
     def self.parse(identifier)
+      unless identifier.is_a?(String)
+        raise Pubid::Errors::InvalidInputError,
+              Pubid::INPUT_NOT_A_STRING_MESSAGE
+      end
+
       if identifier.length > Pubid::MAX_INPUT_LENGTH
-        raise ArgumentError, Pubid::INPUT_TOO_LONG_MESSAGE
+        raise Pubid::Errors::InvalidInputError, Pubid::INPUT_TOO_LONG_MESSAGE
       end
 
       # Use the Parser class's preprocessing method
@@ -87,36 +93,36 @@ module Pubid
     end
 
     # Per-flavor format registry: inherits global formats, overrides :human
-    Identifiers::Base.format_registry = FormatRegistry.new(parent: ::Pubid::Identifier.format_registry)
-    Identifiers::Base.format_registry.register(:human, renderer: Nist::Renderer)
+    Identifier.format_registry = FormatRegistry.new(parent: ::Pubid::Identifier.format_registry)
+    Identifier.format_registry.register(:human, renderer: Nist::Renderer)
 
     # All identifier classes for external consumption (export, website).
-    # Identifiers::Base is the fallback for unmapped series and is excluded
+    # Identifier is the fallback for unmapped series and is excluded
     # from external type listings.
     # @return [Array<Class>] identifier classes
     def self.identifier_types
-      IDENTIFIER_TYPES.reject { |klass| klass == Identifiers::Base }
+      IDENTIFIER_TYPES.reject { |klass| klass == Identifier }
     end
 
     # Aggregate TYPED_STAGES from all identifier classes.
-    # Identifiers::Base is excluded because it has no typed stages and acts
+    # Identifier is excluded because it has no typed stages and acts
     # as the fallback for unmapped series.
     # @return [Array<Pubid::Components::TypedStage>] all typed stages
     def self.all_typed_stages
       @all_typed_stages ||= identifier_types
-        .reject { |klass| klass == Identifiers::Base }
+        .reject { |klass| klass == Identifier }
         .flat_map(&:typed_stages)
         .freeze
     end
 
     # Lookup: type code -> identifier class.
-    # Identifiers::Base is excluded so it never gets selected by type code.
+    # Identifier is excluded so it never gets selected by type code.
     # @param code [String, Symbol] the type code to find
     # @return [Class, nil] the matching identifier class, or nil
     def self.locate_type(code)
       type_str = code.to_s
       identifier_types
-        .reject { |klass| klass == Identifiers::Base }
+        .reject { |klass| klass == Identifier }
         .find { |klass| klass.typed_stages.any? { |ts| ts.type_code.to_s == type_str } }
     end