relaton-iso 2.1.5 → 2.2.0.pre.alpha.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 36e773a9fdc9851a0b9efa4fdbbcf9d34c22ad2cb55ac34fb1af86974273a48c
-  data.tar.gz: 587a6518297383f705e02cf9db43f28fa3824c6129b379e25fd5aaaa24c9db91
+  metadata.gz: 162f2a0536cbb7c7d32b893c2acd75049c9297660fc8a4873459a60f36b0fbff
+  data.tar.gz: 344fae9765adff7581674eb8e03b64a1402667989771306290e650e5f408d1f0
 SHA512:
-  metadata.gz: 58c4fe78b4a8b42272a6d24f3ad72f4e3308d0536913ab910f60b734aeaea2dea7f036d60733923566998bef6e1b2980fa718aea3fcdf7e08d97475876e7a833
-  data.tar.gz: f269f06947f7051219053f4247a7c1bc664b46cbd430e43567d5a76f6c94371a5224c26ff93102380ceb42100d96b618b168a51319847b0203d3fe875ce5582c
+  metadata.gz: 285cdea0029f25e27cac57b01fa54f35a5a1fe50d1cd16c8537e0a04832bce769ea6fe66ab1ae0b576b7ed0fc005e7f755a065dc0b110212ba4921446f75dd5f
+  data.tar.gz: e27586d77e4862bfb98847e03d9087f51931de94c88b46bc9f3c757c3e0740fac6848b95e949806cb9b1f4749a02cd677bb250ca97f38d346b8c3a4f312a8806

data/Gemfile

@@ -5,6 +5,13 @@ git_source(:github) { |repo_name| "https://github.com/#{repo_name}" }
 # Specify your gem's dependencies in relaton_iso.gemspec
 gemspec
 
+# Use local monorepo sibling gems where available.
+Dir["../*/"].each do |dir|
+  name = File.basename(dir)
+  next if name == File.basename(__dir__)
+  next unless File.exist?(File.join(dir, "#{name}.gemspec"))
+  gem name, path: dir
+end
 
 gem "byebug"
 gem "equivalent-xml"

data/Rakefile

@@ -11,8 +11,8 @@ namespace :spec do
     require "net/http"
     require "uri"
 
-    url = "https://raw.githubusercontent.com/relaton/relaton-data-iso/v2/index-v1.zip"
-    dest = File.join(__dir__, "spec", "fixtures", "index-v1.zip")
+    url = "https://raw.githubusercontent.com/relaton/relaton-data-iso/v2/index-v2.zip"
+    dest = File.join(__dir__, "spec", "fixtures", "index-v2.zip")
 
     puts "Downloading #{url} ..."
     uri = URI.parse(url)

data/lib/relaton/iso/bibliography.rb

@@ -40,8 +40,10 @@ module Relaton
         # opts[:all_parts] ||= $~ && opts[:all_parts].nil?
 
         query_pubid = ::Pubid::Iso::Identifier.parse(code)
-        query_pubid.root.year = year.to_i if year&.respond_to?(:to_i)
-        query_pubid.root.all_parts ||= opts[:all_parts]
+        if year&.respond_to?(:to_i)
+          query_pubid.root.date = ::Pubid::Components::Date.new(year: year.to_s)
+        end
+        query_pubid.root.all_parts = opts[:all_parts] if opts[:all_parts]
         Util.info "Fetching from Relaton repository ...", key: query_pubid.to_s
 
         hits, missed_year_ids = isobib_search_filter(query_pubid, opts)
@@ -57,7 +59,7 @@ module Relaton
 
         response_pubid = ret.docidentifier.find(&:primary) # .sub(" (all parts)", "")
         Util.info "Found: `#{response_pubid}`", key: query_pubid.to_s
-        get_all = (query_pubid.root.year && opts[:keep_year].nil?) || opts[:keep_year] || opts[:all_parts] ||
+        get_all = (query_pubid.root.date&.year && opts[:keep_year].nil?) || opts[:keep_year] || opts[:all_parts] ||
           opts[:publication_date_before] || opts[:publication_date_after]
         if get_all
           filter_item_by_date(ret, opts) if date_filter
@@ -65,7 +67,7 @@ module Relaton
         end
 
         ret.to_most_recent_reference
-      rescue ::Pubid::Core::Errors::ParseError
+      rescue Parslet::ParseFailed
         Util.warn "Is not recognized as a standards identifier.", key: code
         nil
       end
@@ -95,7 +97,7 @@ module Relaton
 
         query_pubid.publisher == pubid.publisher &&
           query_pubid.number == pubid.number &&
-          query_pubid.copublisher == pubid.copublisher &&
+          query_pubid.copublishers == pubid.copublishers &&
           (any_types_stages || query_pubid.stage == pubid.stage) &&
           (any_types_stages || query_pubid.is_a?(pubid.class))
       end
@@ -109,10 +111,12 @@ module Relaton
 
         # filter by year
         hit_collection.select! do |hit|
-          hit.pubid.year ||= hit.hit[:year]
+          if hit.pubid.date&.year.nil? && hit.hit[:year]
+            hit.pubid.date = ::Pubid::Components::Date.new(year: hit.hit[:year].to_s)
+          end
           next true if check_year(year, hit)
 
-          missed_year_ids << hit.pubid.to_s if hit.pubid.year
+          missed_year_ids << hit.pubid.to_s if hit.pubid.date&.year
           false
         end
 
@@ -195,7 +199,7 @@ module Relaton
       # @param hit [Relaton::Iso::Hit]
       # @return [Integer]
       def hit_year(hit)
-        yr = hit.pubid&.year || hit.hit[:year] || hit.pubid&.root&.year
+        yr = hit.pubid&.date&.year || hit.hit[:year] || hit.pubid&.root&.date&.year
         yr.to_i
       end
 
@@ -253,9 +257,14 @@ module Relaton
       end
 
       def check_year(year, hit) # rubocop:disable Metrics/AbcSize
-        (hit.pubid.base.nil? && hit.pubid.year.to_s == year.to_s) ||
-          (!hit.pubid.base.nil? && hit.pubid.base.year.to_s == year.to_s) ||
-          (!hit.pubid.base.nil? && hit.pubid.year.to_s == year.to_s)
+        pub = hit.pubid
+        own_year = pub.date&.year.to_s
+        base_year = pub.base_identifier&.date&.year.to_s
+        if pub.base_identifier.nil?
+          own_year == year.to_s
+        else
+          base_year == year.to_s || own_year == year.to_s
+        end
       end
 
       # @param pubid [Pubid::Iso::Identifier] PubID with no results
@@ -264,7 +273,7 @@ module Relaton
 
         if missed_year_ids.any?
           ids = missed_year_ids.map { |i| "`#{i}`" }.join(", ")
-          Util.info "TIP: No match for edition year #{pubid.year}, but matches exist for #{ids}.", key: pubid.to_s
+          Util.info "TIP: No match for edition year #{pubid.date&.year}, but matches exist for #{ids}.", key: pubid.to_s
         end
 
         if tip_ids.any?
@@ -276,7 +285,7 @@ module Relaton
           Util.info "TIP: If it cannot be found, the document may no longer be published in parts.", key: pubid.to_s
         else
           Util.info "TIP: If you wish to cite all document parts for the reference, " \
-                    "use `#{pubid.to_s(format: :ref_undated)} (all parts)`.", key: pubid.to_s
+                    "use `#{pubid.exclude(:date)} (all parts)`.", key: pubid.to_s
         end
 
         nil
@@ -326,12 +335,17 @@ module Relaton
             !(query_pubid.root.all_parts && i.pubid.part.nil?)
         end
 
-        filter_hits_by_year(hit_collection, query_pubid.root.year)
+        filter_hits_by_year(hit_collection, query_pubid.root.date&.year)
       end
 
       def build_excludings(all_parts, any_types_stages)
-        excludings = %i[year edition all_parts]
-        excludings += %i[type stage iteration] if any_types_stages
+        # 2.x attribute names: :year → :date, :iteration → :stage_iteration.
+        # Always exclude :typed_stage: parse fills the default-published
+        # typed_stage with original_abbr="" while .create leaves it nil,
+        # so equality would never hold against indexed/created rows
+        # otherwise.
+        excludings = %i[date edition all_parts typed_stage]
+        excludings += %i[type stage stage_iteration] if any_types_stages
         excludings << :part if all_parts
         excludings
       end
@@ -340,10 +354,28 @@ module Relaton
         if pubid.is_a? String then pubid == query_pubid.to_s
         else
           pubid = pubid.dup
-          pubid.base = pubid.base.exclude(:year, :edition) if pubid.base
-          pubid.exclude(*excludings) == no_year_ref
+          pubid.base_identifier = pubid.base_identifier.exclude(:date, :edition) if pubid.base_identifier
+          normalize_compound_part(pubid.exclude(*excludings)) == no_year_ref
         end
       end
+
+      # @TODO TEMP WORKAROUND (pubid 2.x migration): the v1-generated index
+      # stores a compound part such as "5-1-3" in :part with no :subpart, and
+      # Relaton::Index builds each row via Pubid::Iso::Identifier.from_hash(id),
+      # which keeps it as part="5-1-3" subpart=nil. A parsed query (no_year_ref)
+      # splits it (part="5", subpart="1-3"), so the two never compare equal.
+      # Re-split the candidate's compound part on the first dash to mirror parse
+      # before comparing. `exclude` returns a fresh instance, so mutating this
+      # copy is safe. Remove once pubid create() splits compound parts itself.
+      def normalize_compound_part(pubid)
+        num = pubid.part&.value.to_s
+        return pubid unless pubid.subpart.nil? && num.include?("-")
+
+        head, tail = num.split("-", 2)
+        pubid.part = ::Pubid::Iso::Components::Code.new(value: head)
+        pubid.subpart = ::Pubid::Iso::Components::Code.new(value: tail)
+        pubid
+      end
     end
   end
 end

data/lib/relaton/iso/data_fetcher.rb

@@ -12,11 +12,21 @@ module Relaton
     # (see https://www.iso.org/open-data.html) and write each one as a YAML
     # file under `@output`.
     #
-    # `source` modes (matching the `Relaton::Core::DataFetcher.fetch` arg):
+    # The upstream feed has no delta API, so any run that proceeds re-downloads
+    # and re-ingests the whole feed. There is therefore no value in a partial
+    # update: a run either skips entirely or does a full replace. `source` modes
+    # (matching the `Relaton::Core::DataFetcher.fetch` arg):
     #
-    # * `"iso-open-data"` (default) - skip the run if the upstream
-    #   `Last-Modified` header matches `LAST_MODIFIED_FILE`.
-    # * `"iso-open-data-all"` - clear `@output` and re-emit every record.
+    # * `"iso-open-data"` (default) - skip when the feed's `Last-Modified` is
+    #   unchanged; otherwise wipe `@output` + index and rebuild from scratch.
+    # * `"iso-open-data-all"` - the same full rebuild, but ignore the
+    #   `Last-Modified` short-circuit and always run.
+    #
+    # Wiping happens here, after the short-circuit decision, so `@output` and the
+    # index always mirror the current feed (records that have left it don't
+    # linger as stale files or dangling index entries) without risking an empty
+    # tree on a skipped run. `#fetch` returns true when it rebuilt, false when
+    # it skipped, so callers can chain follow-up work (e.g. the pubid-v1 index).
     #
     class DataFetcher < Core::DataFetcher
       OPEN_DATA_URL = "https://isopublicstorageprod.blob.core.windows.net/" \
@@ -45,9 +55,9 @@ module Relaton
 
         Util.info "Fetching ISO Open Data (mode: #{@source})..."
         last_modified = fetch_last_modified
-        return if up_to_date?(last_modified)
+        return false if up_to_date?(last_modified)
 
-        prepare_output
+        reset_output
         jsonl_path = download_dataset
         ref_index, amend_index, date_index = build_ref_index(jsonl_path)
         tc_index = build_tc_index
@@ -57,6 +67,7 @@ module Relaton
         index.save
         save_last_modified(last_modified)
         report_errors
+        true
       rescue StandardError => e
         Util.error "#{e.message}\n#{e.backtrace.join("\n")}"
         raise
@@ -103,8 +114,13 @@ module Relaton
         File.write(LAST_MODIFIED_FILE, last_modified, encoding: "UTF-8")
       end
 
-      def prepare_output
-        FileUtils.rm_rf(@output) if @full_refresh
+      # Reset the data tree and the index together so the rebuild is a clean
+      # mirror of the feed. Called only after the short-circuit, so a skipped run
+      # never strands an empty tree. `Core::DataFetcher.fetch` recreates the
+      # directory before ingest writes into it.
+      def reset_output
+        FileUtils.rm_rf(@output)
+        index.remove_all
         FileUtils.mkdir_p(@output)
       end
 
@@ -174,9 +190,9 @@ module Relaton
 
       def amend_base(ref)
         pubid = ::Pubid::Iso::Identifier.parse(ref)
-        return nil unless pubid.respond_to?(:base) && pubid.base
+        return nil unless pubid.base_identifier
 
-        pubid.base.to_s
+        pubid.base_identifier.to_s
       rescue StandardError
         nil
       end
@@ -260,10 +276,42 @@ module Relaton
 
       def write_file(file, doc, docid)
         @files << file
-        index.add_or_update(docid.pubid || docid.content.to_s, file)
+        index_primary(docid, file)
         File.write(file, serialize(doc), encoding: "UTF-8")
       end
 
+      # Add a document's primary id to the index. With pubid 2.x every ISO id
+      # is expected to parse; if one does not (`docid.pubid` is nil) record it
+      # so `report_errors` raises a tracked GitHub issue at the end, and skip
+      # the index entry rather than indexing a raw string (which would crash
+      # the index sort: `get_id_number` calls `.number` on the id). The data
+      # file is still written, so the document is not lost — only unindexed
+      # until its id parses.
+      def index_primary(docid, file)
+        unless docid.pubid
+          unparseable_ids << [docid.content.to_s, file]
+          return
+        end
+        index.add_or_update(docid.pubid, file)
+      end
+
+      def unparseable_ids
+        @unparseable_ids ||= []
+      end
+
+      # Surface unparseable primary ids through the shared error-reporting
+      # machinery (a "Error fetching documents" GitHub issue in CI) so they are
+      # visible and tracked, not silently dropped in the action log. The
+      # gh_issue logger channel is registered inside `report_errors`, so emit
+      # these at :error after it is set up and before `super` creates the issue.
+      def report_errors
+        gh_issue
+        unparseable_ids.each do |content, file|
+          log_error "Unparseable primary id `#{content}` was not indexed (#{file})"
+        end
+        super
+      end
+
       # --- static merge -----------------------------------------------------
 
       def merge_static_files
@@ -274,7 +322,7 @@ module Relaton
           did = item.docidentifier.detect(&:primary)
           next unless did
 
-          index.add_or_update(did.pubid || did.content.to_s, f)
+          index_primary(did, f)
         end
       end
 

data/lib/relaton/iso/data_parser.rb

@@ -37,7 +37,6 @@ module Relaton
         "Cor" => "technical-corrigendum",
         "Add" => "addendum",
         "Suppl" => "supplement",
-        "Ext" => "extract",
       }.freeze
 
       DOC_URL = "https://www.iso.org/standard/%d.html"
@@ -123,8 +122,9 @@ module Relaton
       end
 
       def iso_reference_pubid
-        params = pubid.to_h.except(:typed_stage)
-        ::Pubid::Iso::Identifier.create(language: "en", **params)
+        pubid.dup.tap do |id|
+          id.languages = [::Pubid::Components::Language.new(code: "en", original_code: "E")]
+        end
       rescue StandardError
         nil
       end
@@ -133,12 +133,10 @@ module Relaton
         return @urn_pubid if defined?(@urn_pubid)
 
         @urn_pubid = begin
-          dup_pubid = pubid.dup
-          if dup_pubid.respond_to?(:stage=) && stage_dotted &&
-             dup_pubid.respond_to?(:stage) && dup_pubid.stage.nil?
-            dup_pubid.stage = ::Pubid::Iso::Identifier.parse_stage(stage_dotted)
-          end
-          dup_pubid
+          # Override stage even when the parsed pubid carries the default
+          # "published" stage — relaton's currentStage (e.g. 9092 = Withdrawn)
+          # is the authoritative source for URN stage.
+          stage_dotted ? pubid.with_harmonized_stage(stage_dotted) : pubid.dup
         rescue StandardError
           nil
         end
@@ -386,9 +384,9 @@ module Relaton
       end
 
       def base_relation
-        return [] unless pubid&.respond_to?(:base) && pubid.base
+        return [] unless pubid&.base_identifier
 
-        [relation_for(pubid.base.to_s, "updates")]
+        [relation_for(pubid.base_identifier.to_s, "updates")]
       end
 
       def relation_for(ref, type)

data/lib/relaton/iso/hit.rb

@@ -47,7 +47,7 @@ module Relaton
 
       def create_pubid(id)
         if id.is_a?(Hash)
-          ::Pubid::Iso::Identifier.create(**id)
+          ::Pubid::Iso::Identifier.from_hash(id)
         else
           id
         end

data/lib/relaton/iso/hit_collection.rb

@@ -12,15 +12,35 @@ module Relaton
         @opts ||= {}
       end
 
+      # Maps the legacy 1.x exclude symbols to their pubid 2.x attribute
+      # names. The public excludings API still uses :year/:iteration for
+      # backwards compatibility with existing call sites and specs.
+      LEGACY_EXCLUDE_MAP = { year: :date, iteration: :stage_iteration }.freeze
+      private_constant :LEGACY_EXCLUDE_MAP
+
+      def translate_excludings(attrs)
+        out = attrs.map { |a| LEGACY_EXCLUDE_MAP[a] || a }
+        # Excluding :stage implies excluding :typed_stage too — the two
+        # carry overlapping data and their default-published values can
+        # differ in trivia (e.g. original_abbr "" vs nil), so leaving
+        # typed_stage in the comparison breaks otherwise-equal matches.
+        out << :typed_stage if out.include?(:stage) && !out.include?(:typed_stage)
+        out
+      end
+
       def ref_pubid_no_year
-        @ref_pubid_no_year ||= ref.base ? ref.dup.tap { |r| r.base = r.base.exclude(:year) } : ref.exclude(:year)
+        @ref_pubid_no_year ||=
+          if ref.base_identifier
+            ref.dup.tap { |r| r.base_identifier = r.base_identifier.exclude(:date) }
+          else
+            ref.exclude(:date)
+          end
       end
 
       def ref_pubid_excluded
         return @ref_pubid_excluded if defined? @ref_pubid_excluded
 
-        ref_excludings = excludings.dup
-        ref_excludings << :all_parts
+        ref_excludings = translate_excludings(excludings) + [:all_parts]
         @ref_pubid_excluded ||= ref_pubid_no_year.exclude(*ref_excludings)
       end
 
@@ -30,38 +50,68 @@ module Relaton
       # @return [Array<Relaton::Iso::Hit>] hits
       #
       def find # rubocop:disable Metrics/AbcSize
-        @array = index.search do |row|
-          row[:id].is_a?(Hash) || row[:id].is_a?(::Pubid::Core::Identifier::Base) ? pubid_match?(row[:id]) : ref.to_s(with_prf: true) == row[:id]
+        # Pass `ref` (a Pubid::Identifier, not a String) so the index can
+        # narrow candidates by number via binary search before applying the
+        # block, instead of a full O(n) scan of every row. Every row's `:id`
+        # is already a Pubid::Identifier — Relaton::Index deserialized it via
+        # the `pubid_class` passed in `#index` — so `pubid_match?` compares
+        # Pubid to Pubid directly.
+        @array = index.search(ref) do |row|
+          pubid_match?(row[:id])
         end.map { |row| Hit.new row, self }
-          .sort_by! { |h| h.pubid.to_s }
-          .reverse!
+        # An all-parts query drops :part from the match, so multiple rows can
+        # resolve to the same pubid; collapse them so each part appears once.
+        @array.uniq! { |h| h.pubid.to_s } if ref.root.all_parts
+        # Most-recent first (pubid string desc ~ year desc), then float
+        # published-stage ids above drafts. An undated query excludes :stage
+        # when matching, so a future draft (e.g. ISO/AWI) matches alongside the
+        # published edition; without this the draft would sort first lexically
+        # ("ISO/AWI …" > "ISO …") and be returned by fetch_doc's `first`. The
+        # index id carries no lifecycle status, so the parsed stage is the only
+        # signal available here. partition is stable, preserving the year order
+        # within each group.
+        @array.sort_by! { |h| h.pubid.to_s }.reverse!
+        published, drafts = @array.partition do |h|
+          h.pubid && default_published_stage?(h.pubid)
+        end
+        @array = published + drafts
         self
       end
 
-      def pubid_match?(id)
-        pubid = create_pubid(id)
-        return false unless pubid
-
-        # pubid.base = pubid.base.exclude(:year, :edition) if pubid.base
-        dir_excludings = excludings.dup
-        dir_excludings << :edition unless pubid.typed_stage_abbrev == "DIR"
-        exclude_id_attrs(pubid, *dir_excludings) == ref_pubid_excluded
+      def pubid_match?(pubid)
+        match_excludings = translate_excludings(excludings) + [:all_parts]
+        match_excludings << :edition unless pubid.typed_stage&.abbr&.include?("DIR")
+        # Only the candidate is built via .create (from the index) and so may
+        # carry a compound part; `ref_pubid_no_year` is always a parsed pubid,
+        # already split, so it needs no normalization.
+        cand = normalize_compound_part(exclude_id_attrs(pubid, *match_excludings))
+        cand == exclude_id_attrs(ref_pubid_no_year, *match_excludings)
       end
 
-      def create_pubid(id)
-        return id if id.is_a?(::Pubid::Core::Identifier::Base)
-
-        ::Pubid::Iso::Identifier.create(**id)
-      rescue StandardError => e
-        Util.warn e.message, key: ref.to_s
+      # @TODO TEMP WORKAROUND (pubid 2.x migration): the v1-generated index
+      # stores a compound part such as "5-1-3" in :part with no :subpart, and
+      # Relaton::Index builds each row via Pubid::Iso::Identifier.from_hash(id),
+      # which keeps it as part="5-1-3" subpart=nil. A parsed query splits it
+      # (part="5", subpart="1-3"), so the two never compare equal. Re-split the
+      # compound part on the first dash to mirror parse before comparing.
+      # `exclude` returns a fresh instance, so mutating this copy is safe.
+      # Remove once pubid create() splits compound parts itself.
+      def normalize_compound_part(pubid)
+        num = pubid.part&.value.to_s
+        return pubid unless pubid.subpart.nil? && num.include?("-")
+
+        head, tail = num.split("-", 2)
+        pubid.part = ::Pubid::Iso::Components::Code.new(value: head)
+        pubid.subpart = ::Pubid::Iso::Components::Code.new(value: tail)
+        pubid
       end
 
       def exclude_id_attrs(pubid, *attrs)
         xid = pubid.exclude(*attrs)
         curr = xid
-        while curr.base
-          curr.base = curr.base.exclude(*attrs)
-          curr = curr.base
+        while curr.base_identifier
+          curr.base_identifier = curr.base_identifier.exclude(*attrs)
+          curr = curr.base_identifier
         end
         xid
       end
@@ -71,23 +121,26 @@ module Relaton
 
         excl_attrs = %i[year]
         excl_attrs << :part if ref.root.part.nil? || ref.root.all_parts
-        if ref.stage.nil? || ref.root.all_parts
+        if default_published_stage?(ref) || ref.root.all_parts
           excl_attrs << :stage
           excl_attrs << :iteration
         end
-        # excl_parts << :edition if ref.root.edition.nil? || all_parts
         @excludings = excl_attrs
       end
 
+      # Pubid 2.x auto-populates a published-stage default on parse/.create,
+      # so ref.stage is never nil. Treat that default as "no stage specified".
+      def default_published_stage?(pubid)
+        return true if pubid.typed_stage.nil?
+
+        pubid.typed_stage.stage_code.to_s == "published"
+      end
+
       def index
         @index ||= Relaton::Index.find_or_create(
           :iso,
           url: "#{ENDPOINT}#{INDEXFILE}.zip",
           file: "#{INDEXFILE}.yaml",
-          id_keys: %i[publisher number copublisher part year edition type stage
-                      iteration joint_document tctype sctype wgtype tcnumber
-                      scnumber wgnumber dirtype base supplements addendum
-                      jtc_dir month amendments corrigendums language],
           pubid_class: ::Pubid::Iso::Identifier,
         )
       end
@@ -107,9 +160,9 @@ module Relaton
       def to_all_parts # rubocop:disable Metrics/AbcSize,Metrics/CyclomaticComplexity
         parts = @array.select { |h| h.pubid.part }
         if opts[:publication_date_before] || opts[:publication_date_after]
-          parts = parts.select { |h| Bibliography.send(:year_in_range?, (h.pubid.year || h.hit[:year]).to_i, opts) }
+          parts = parts.select { |h| Bibliography.send(:year_in_range?, (h.pubid.date&.year || h.hit[:year]).to_i, opts) }
         end
-        hit = parts.min_by { |h| h.pubid.part.to_i }
+        hit = parts.min_by { |h| h.pubid.part.value.to_i }
         return @array.first&.item unless hit
 
         bibitem = hit.item

data/lib/relaton/iso/item_data.rb

@@ -21,7 +21,7 @@ module Relaton
 
       def create_id_from_pubid(content, without_date)
         pubid = without_date ? content.exclude(:year) : content
-        self.id = pubid.to_s(with_prf: true).gsub(/\W+/, "")
+        self.id = pubid.to_s.gsub(/\W+/, "")
       end
     end
   end

data/lib/relaton/iso/model/docidentifier.rb

@@ -27,7 +27,7 @@ module Relaton
         else
           parsed =
             case value
-            when ::Pubid::Iso::Identifier::Base then value
+            when ::Pubid::Iso::Identifier then value
             when String
               begin
                 ::Pubid::Iso::Identifier.parse(value)
@@ -42,6 +42,13 @@ module Relaton
 
           if parsed
             @pubid = parsed
+            # TC committee documents have a canonical spelling ("… N1110")
+            # that pubid renders with a space ("… N 1110"). Preserve the
+            # source string (same intent as the iso-tc bypass) while keeping
+            # the parsed pubid for any structural operations.
+            if value.is_a?(String) && parsed.is_a?(::Pubid::Iso::Identifiers::TcDocument)
+              @raw_content = value
+            end
           elsif value.is_a?(String)
             @raw_content = value
           end
@@ -80,18 +87,18 @@ module Relaton
       end
 
       def remove_date!
-        remove_attr!(:year)
+        remove_attr!(:date)
       end
 
       def exclude_year
         return @raw_content if @raw_content
         return nil unless @pubid
 
-        pubid = @pubid.exclude(:year)
+        pubid = @pubid.exclude(:date)
         current = pubid
-        while current.base
-          current.base = current.base.exclude(:year)
-          current = current.base
+        while current.base_identifier
+          current.base_identifier = current.base_identifier.exclude(:date)
+          current = current.base_identifier
         end
         pubid
       end
@@ -100,11 +107,10 @@ module Relaton
 
       def render_pubid(pubid)
         case type
-        when "URN" then pubid.urn
-        when "iso-reference", "iso-with-lang"
-          pubid.to_s(format: :ref_num_short, with_prf: true)
+        when "URN" then pubid.to_urn
+        when "ISO" then pubid.exclude(:languages).to_s
         else
-          pubid.to_s(with_prf: true)
+          pubid.to_s
         end
       end
 
@@ -112,10 +118,10 @@ module Relaton
         return unless @pubid
 
         @pubid.send("#{attr}=", nil)
-        base = @pubid.base
+        base = @pubid.base_identifier
         while base
           base.send("#{attr}=", nil)
-          base = base.base
+          base = base.base_identifier
         end
         refresh_content!
       end

data/lib/relaton/iso/model/doctype.rb

@@ -4,7 +4,7 @@ module Relaton
       TYPES = %w[
         international-standard technical-specification technical-report publicly-available-specification
         international-workshop-agreement guide recommendation amendment technical-corrigendum directive
-        committee-document addendum supplement extract
+        committee-document addendum supplement
       ].freeze
 
       attribute :content, :string, values: TYPES

data/lib/relaton/iso/scraper.rb

@@ -104,16 +104,14 @@ module Relaton
         return @pubid if @pubid
 
         @pubid = ::Pubid::Iso::Identifier.parse(id)
-        @pubid.root.edition ||= edition.content if @pubid.base
+        @pubid.root.edition ||= edition.content if @pubid.base_identifier
         @pubid
       rescue StandardError => e
         Util.error "Failed to parse pubid from #{id}: #{e.message}"
       end
 
       def urn
-        pubid_dup = pubid.dup
-        pubid_dup.stage ||= ::Pubid::Iso::Identifier.parse_stage(stage_code)
-        pubid_dup
+        pubid.with_harmonized_stage(stage_code)
       end
 
       def edition
@@ -143,8 +141,9 @@ module Relaton
       # @return [String] English reference identifier
       #
       def isoref
-        params = pubid.to_h.except(:typed_stage)
-        ::Pubid::Iso::Identifier.create(language: "en", **params).to_s(format: :ref_num_short)
+        pubid.dup.tap do |id|
+          id.languages = [::Pubid::Components::Language.new(code: "en", original_code: "E")]
+        end.to_s
       end
 
       private

data/lib/relaton/iso/type/pubid.rb

@@ -1,7 +1,7 @@
 module Relaton
   module Iso
     module Type
-      # Lutaml-model attribute type that preserves `Pubid::Iso::Identifier::Base`
+      # Lutaml-model attribute type that preserves `Pubid::Iso::Identifier`
       # instances on the way in and stringifies them on the way out.
       #
       # The default `:string` type calls `.to_s` during `cast`, which loses the

data/lib/relaton/iso/version.rb

@@ -2,6 +2,6 @@
 
 module Relaton
   module Iso
-    VERSION = "2.1.5"
+    VERSION = "2.2.0.pre.alpha.1"
   end
 end

data/lib/relaton/iso.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 
 require "net/http"
-require "pubid/iso"
+require "pubid"
 require "relaton/index"
 require "isoics"
 require "relaton/bib"
@@ -18,7 +18,7 @@ require_relative "iso/bibliography"
 
 module Relaton
   module Iso
-    INDEXFILE = "index-v1"
+    INDEXFILE = "index-v2"
 
     def self.grammar_hash
       # gem_path = File.expand_path "..", __dir__

data/relaton-iso.gemspec

@@ -24,11 +24,11 @@ Gem::Specification.new do |spec|
   spec.bindir        = "exe"
   spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
   spec.require_paths = ["lib"]
-  spec.required_ruby_version = Gem::Requirement.new(">= 3.2.0")
+  spec.required_ruby_version = Gem::Requirement.new(">= 3.3.0")
 
   spec.add_dependency "isoics", "~> 0.1.6"
-  spec.add_dependency "pubid-iso", "~> 1.15.20"
-  spec.add_dependency "relaton-bib", "~> 2.1.0"
-  spec.add_dependency "relaton-core", "~> 0.0.12"
-  spec.add_dependency "relaton-index", "~> 0.2.12"
+  spec.add_dependency "pubid", "~> 2.0.0.pre.alpha.3"
+  spec.add_dependency "relaton-bib", "~> 2.2.0.pre.alpha.1"
+  spec.add_dependency "relaton-core", "~> 2.2.0.pre.alpha.1"
+  spec.add_dependency "relaton-index", "~> 2.2.0.pre.alpha.1"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: relaton-iso
 version: !ruby/object:Gem::Version
-  version: 2.1.5
+  version: 2.2.0.pre.alpha.1
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-06-27 00:00:00.000000000 Z
+date: 2026-06-26 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: isoics
@@ -25,61 +25,61 @@ dependencies:
       - !ruby/object:Gem::Version
         version: 0.1.6
 - !ruby/object:Gem::Dependency
-  name: pubid-iso
+  name: pubid
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 1.15.20
+        version: 2.0.0.pre.alpha.3
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 1.15.20
+        version: 2.0.0.pre.alpha.3
 - !ruby/object:Gem::Dependency
   name: relaton-bib
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 2.1.0
+        version: 2.2.0.pre.alpha.1
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 2.1.0
+        version: 2.2.0.pre.alpha.1
 - !ruby/object:Gem::Dependency
   name: relaton-core
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.0.12
+        version: 2.2.0.pre.alpha.1
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.0.12
+        version: 2.2.0.pre.alpha.1
 - !ruby/object:Gem::Dependency
   name: relaton-index
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.2.12
+        version: 2.2.0.pre.alpha.1
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.2.12
+        version: 2.2.0.pre.alpha.1
 description: 'Relaton::Iso: retrieve ISO Standards for bibliographic use using the
   IsoBibliographicItem model'
 email:
@@ -93,7 +93,6 @@ files:
 - ".gitignore"
 - ".hound.yml"
 - ".rspec"
-- ".rubocop.yml"
 - CLAUDE.md
 - CODE_OF_CONDUCT.md
 - Gemfile
@@ -143,7 +142,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 3.2.0
+      version: 3.3.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="

data/.rubocop.yml

@@ -1,12 +0,0 @@
-# This project follows the Ribose OSS style guide.
-# https://github.com/riboseinc/oss-guides
-# All project-specific additions and overrides should be specified in this file.
-
-require: rubocop-rails
-
-inherit_from:
-  - https://raw.githubusercontent.com/riboseinc/oss-guides/master/ci/rubocop.yml
-AllCops:
-  TargetRubyVersion: 3.2
-Rails:
-  Enabled: false