pubid 2.0.0.pre.alpha.44 → 2.0.0.pre.alpha.46

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 217d5576587e97055e8ad7cb387e62925495649c4f2d4193692e8e727fac07cf
-  data.tar.gz: a923fa20acab08537ed36a7ce823425fd130903a5a351280a212fb8db577ccae
+  metadata.gz: 8e59d6b8d85efbedab5ee3e65fb772d579e53b08df3a5b1fcea783c9e70c005e
+  data.tar.gz: 8a757f8f946b2998e984ed915bd60896c14a93f4bbce4368810faef5f70b13e6
 SHA512:
-  metadata.gz: 4cbd63b039f15863cdc92000e6a417faad9e050d31194026c757e730bc94789ee050fff50039a9956573425bd93090eb0c9cd90b2753062cd3395e6ea23bd3a9
-  data.tar.gz: d65ce7950395d4ca5277f2396795c88197eb3aed72228ac0d3a55f87282b14c6b5cc233a4e9332cae1637f36d253dce5db63b3e5ac7e0b7822f60ba20ca0ae15
+  metadata.gz: 5c03453929b1941309d16b6b87c8bee8dc54594c80a42336b1af5a986197ec6a5ab253451fed9198a67d00337149577d6cae57ed9e727f6a0a22fc26ac42480e
+  data.tar.gz: ae7af0fd7a05bb2879d50107ee5d3c1f617e26d2758f0a9a75f28b9ccb124cb914478fe88ec2a3ec64018ab7f8fc391b24e39ab2bfa5905c80fc4e965e6b7d4a

data/lib/pubid/cen_cenelec.rb

@@ -195,7 +195,21 @@ module Pubid
     # @return [Pubid::Components::TypedStage, nil] the matching typed stage
     def self.locate_stage(abbr)
       abbr_str = abbr.to_s.upcase
-      all_typed_stages.find { |s| s.abbr.any? { |a| a.to_s.upcase == abbr_str } }
+      all_typed_stages.find { |s| s.abbr.any? { |a| a.to_s.upcase == abbr_str } } ||
+        # A harmonized stage code ("43.20") names the stage the same way an
+        # abbreviation does; consumers query by it (pubid#251), so an
+        # unmatched abbreviation falls through to the code tables.
+        locate_stage_by_harmonized_code(abbr)
+    end
+
+    # Lookup: harmonized stage code -> typed stage ("43.20" -> the FV EN
+    # entry). The codes are the CEN/CENELEC lifecycle ladder; every code a
+    # consumer can query must resolve (pubid#251).
+    # @param code [String, Symbol] the harmonized stage code to find
+    # @return [Pubid::Components::TypedStage, nil] the matching typed stage
+    def self.locate_stage_by_harmonized_code(code)
+      code_str = code.to_s
+      all_typed_stages.find { |s| s.harmonized_stages.to_a.include?(code_str) }
     end
 
     # Lookup: typed-stage code -> typed stage ("pren" -> the prEN entry).

data/lib/pubid/iso/urn_parser.rb

@@ -142,16 +142,31 @@ module Pubid
         elsif TYPED_STAGE_REVERSE_MAP.key?(parts.first&.upcase) ||
             (parts.first&.match?(/^[A-Za-z]+\.\d+$/) &&
              TYPED_STAGE_REVERSE_MAP.key?(parts.first.upcase.split(".").first))
-          stage_abbr = parts.shift.upcase
-          # Check for iteration (WD.2 format)
-          if stage_abbr.include?(".")
-            abbr_part, iteration_part = stage_abbr.split(".")
-            stage_code = TYPED_STAGE_REVERSE_MAP[abbr_part]
-            stage_from_abbr = stage_code
-            stage_iteration = iteration_part.to_i
-          else
-            stage_code = TYPED_STAGE_REVERSE_MAP[stage_abbr]
-            stage_from_abbr = stage_code
+          stage_abbr = parts.first.upcase
+          # The family comes from the abbreviation without its iteration
+          # ("FDAM.3" is still the amd family).
+          family = Pubid::Iso.locate_stage(stage_abbr.split(".").first)&.type_code&.to_s
+          # A supplement-family stage directly before its own supplement
+          # type token ("...:FDAM:amd:1:v1") names the SUPPLEMENT's stage:
+          # consuming it as the base's stage builds the base as a base-less
+          # supplement, which cannot render.
+          supplement_follows =
+            %w[amd cor suppl add].include?(family) &&
+            parts.drop(1).any? do |token|
+              family == SUPPLEMENT_TYPE_MAP[token.to_s.downcase]&.to_s
+            end
+          unless supplement_follows
+            parts.shift
+            # Check for iteration (WD.2 format)
+            if stage_abbr.include?(".")
+              abbr_part, iteration_part = stage_abbr.split(".")
+              stage_code = TYPED_STAGE_REVERSE_MAP[abbr_part]
+              stage_from_abbr = stage_code
+              stage_iteration = iteration_part.to_i
+            else
+              stage_code = TYPED_STAGE_REVERSE_MAP[stage_abbr]
+              stage_from_abbr = stage_code
+            end
           end
         end
 
@@ -397,6 +412,16 @@ module Pubid
         # Build supplements recursively (forward order: first supplement wraps base,
         # second wraps first, etc. -- matching URN left-to-right semantics)
         supplements.each do |supp|
+          # A bare language token with no supplement identity names the BASE
+          # document: wrapping it would build a supplement with no type,
+          # number or base attribute, and a plain standard drops its base —
+          # the whole document was lost, rendering "ISO (FR)".
+          if supp[:type].nil? && supp[:stage].nil? && supp[:number].nil? &&
+              supp[:year].nil? && supp[:languages]
+            base_hash[:languages] = supp[:languages].join("/")
+            next
+          end
+
           supp_hash = {}
 
           if supp[:stage]

data/lib/pubid/nist/CLAUDE.md

@@ -72,3 +72,11 @@ Read this before you change `lib/pubid/nist/` or `spec/pubid/nist/`. The root
   (`"800-38.A.supp.add."` splits with a trailing `""` in JS but not in
   Ruby; keeping it appends a stray `.`). Locked by
   `spec/pubid/nist/urn_parser_spec.rb`.
+
+- **FIPS MR keeps the parsed publisher.** `FederalInformationProcessingStandards`
+  sets `default_publisher` to `""`, and its `to_mr_style` once started from
+  `series_code` alone, so `NIST.FIPS.197-upd1.ipd` rendered `FIPS.197-upd1.ipd`
+  — not the DOI suffix NIST publishes. It now mirrors its own
+  `to_short_style`: the publisher prefixes only when the input carried one
+  (`NIST.`/`NBS.`), and a bare `FIPS 140-3` still renders `FIPS.140-3`.
+  Locked by `spec/pubid/nist/identifiers/fips_spec.rb`.

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

@@ -37,11 +37,12 @@ module Pubid
           ""
         end
 
-        # Override to_mr_style to not add "NIST" prefix for FIPS
+        # Override to_mr_style so FIPS adds no default "NIST" prefix
         def to_mr_style
-          # "FIPS.46e1977" (machine-readable with dots, no publisher prefix)
+          # "NIST.FIPS.197" keeps a parsed publisher (the DOI suffix NIST
+          # publishes); "FIPS.46e1977" has none and invents none.
           # Note: edition is appended directly to number without dot for FIPS
-          result = series_code
+          result = [publisher, series_code].compact.join(".")
           result += ".#{number}" if number
 
           # FIPS uses dash notation for parts; render here and skip the generic

data/lib/pubid/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Pubid
-  VERSION = "2.0.0.pre.alpha.44"
+  VERSION = "2.0.0.pre.alpha.46"
 end

data/lib/tasks/benchmark.rake

@@ -0,0 +1,68 @@
+# frozen_string_literal: true
+
+require "benchmark"
+
+namespace :benchmark do
+  desc "Parse throughput per flavor over the testsuite corpus " \
+       "(PUBID_TESTSUITE_PATH=<dir>, FLAVOR=ieee,iso, LIMIT=N)"
+  task :corpus do
+    require "pubid"
+
+    corpus_dir = ENV["PUBID_TESTSUITE_PATH"] ||
+      File.expand_path("../../../pubid-testsuite/tests", __dir__)
+    unless Dir.exist?(corpus_dir)
+      abort "corpus not found at #{corpus_dir} — set PUBID_TESTSUITE_PATH " \
+            "to a pubid-testsuite checkout's tests/ directory"
+    end
+
+    limit = ENV["LIMIT"]&.to_i
+    dir_to_flavor = { "tgpp" => "3gpp" }
+    flavors = if ENV["FLAVOR"]
+                ENV["FLAVOR"].split(",").map(&:strip)
+              else
+                Dir.children(corpus_dir)
+                  .select { |name| File.directory?(File.join(corpus_dir, name)) && !name.start_with?("_") }
+                  .sort
+              end
+
+    Pubid.eager_load_flavors!
+
+    rows = flavors.filter_map do |dir_name|
+      flavor = (dir_to_flavor[dir_name] || dir_name).to_sym
+      mod = Pubid::Registry.get(flavor)
+      next unless mod
+
+      cases = Pubid::Conformance::Corpus.cases(dir_name, corpus_dir)
+      inputs = cases.filter_map { |c| c.representations&.human || c.input }
+      inputs = inputs.first(limit) if limit && limit.positive?
+      next if inputs.empty?
+
+      failures = 0
+      elapsed = ::Benchmark.realtime do
+        inputs.each do |input|
+          mod.parse(input)
+        rescue StandardError
+          failures += 1
+        end
+      end
+      rate = elapsed.zero? ? 0.0 : inputs.size / elapsed
+      { flavor: dir_name, cases: inputs.size, failures: failures,
+        seconds: elapsed, rate: rate }
+    end
+
+    rows.sort_by! { |row| -row[:rate] }
+    total_cases = rows.sum { |row| row[:cases] }
+    total_seconds = rows.sum { |row| row[:seconds] }
+
+    puts format("%-14s %8s %8s %10s %12s", "flavor", "cases", "failed", "seconds", "parses/sec")
+    rows.each do |row|
+      puts format("%-14s %8d %8d %10.3f %12.1f", row[:flavor], row[:cases],
+                  row[:failures], row[:seconds], row[:rate])
+    end
+    puts format("%-14s %8d %8s %10.3f %12.1f", "TOTAL", total_cases, "",
+                total_seconds, total_seconds.zero? ? 0.0 : total_cases / total_seconds)
+    puts
+    puts "ruby #{RUBY_VERSION}; parsanol #{Gem.loaded_specs['parsanol']&.version}; " \
+         "pubid #{Pubid::VERSION}"
+  end
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: pubid
 version: !ruby/object:Gem::Version
-  version: 2.0.0.pre.alpha.44
+  version: 2.0.0.pre.alpha.46
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-10-20 00:00:00.000000000 Z
+date: 2026-10-21 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: lutaml-model
@@ -1047,6 +1047,7 @@ files:
 - lib/pubid/xsf/renderer.rb
 - lib/pubid/xsf/urn_generator.rb
 - lib/pubid/xsf/urn_parser.rb
+- lib/tasks/benchmark.rake
 - lib/tasks/conformance.rake
 - lib/tasks/docs.rake
 - lib/tasks/export.rake