relaton-jis 2.1.0 → 2.2.0.pre.alpha.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 1a2abb23ea701fb33099e16b02eae904d3de747395444ed8286ccfd6f9078cb3
-  data.tar.gz: 7d161343c30d77222dde2f64b07ff71ecb9584a6cdbd9dd978dad41a323c40da
+  metadata.gz: 7ed2ef5528da1c7c4875ae6ffc9f5e18b83fb1b7ef8844596e2ed4d868010b52
+  data.tar.gz: fbeda1c6f0efa761c24eb2f007461a55d6d25c44b6e00063a776b79774bc57ed
 SHA512:
-  metadata.gz: f86fad4a0b3e2eb96d06d0311313fd02a2ea277aa0f09e93551dd1f7f688ff34366249d52d3e45689b740fb579738d7d55edc6e8a5196fafbc39660862ee5a3d
-  data.tar.gz: 103cff93037c28424ed9e665488d1f0accfe0fa79e9e37e5374d7ee8424a4a70fd638f72257e8d35b279c17585071d5523fd42e2bc4887fb60443a9cd19d81df
+  metadata.gz: b5d13203270407ddbdc2b406226cff9955d3f699d9a245cd532b21cd44e0178281ab92c08bbaef27f478be3c77e20f573b2531d213f041aaf97f0994a6275f24
+  data.tar.gz: 2208a5a3d65b6d4223b3a334515b95a05658345e82719d9b62fb8b885dd935fb2a624a6ded1513a4b5680dfb136c9335aea107777288443da360fea8bd77512a

data/CLAUDE.md

@@ -58,4 +58,6 @@ All classes live under `lib/relaton/jis/`:
 
 ## Testing
 
-- **Index fixture:** `spec/fixtures/index-v1.zip` is pre-loaded into `Relaton::Index` pool in `before(:suite)` (configured in `spec/support/webmock.rb`). Run `rake spec:update_index` to refresh from relaton-data-jis.
+- **Index fixture:** `spec/fixtures/index-v2.zip` (the pubid-based `index-v2`) is deserialized via `Pubid::Jis::Identifier` and pre-loaded into the `Relaton::Index` pool in `before(:suite)` (configured in `spec/support/webmock.rb`). Run `rake spec:update_index` to refresh from relaton-data-jis.
+- **Lookup path:** `Bibliography.get`/`search` parse the reference with `pubid`, then `HitCollection` matches against `index-v2` (`Hit#matches?` compares type/series/number/part; year is filtered separately). The `DataFetcher` still generates `index-v1` too, but the runtime no longer reads it.
+- **pubid dependency:** the gemspec requires `pubid ~> 2.0.0.pre.alpha.3`, the published release carrying the working lutaml-model `from_hash` (the earlier `2.0.0.pre.alpha.2` was broken). Override to a local checkout with `bundle config set local.pubid /path/to/pubid`.

data/Gemfile

@@ -5,6 +5,14 @@ source "https://rubygems.org"
 # Specify your gem's dependencies in relaton_jis.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 "rake", "~> 13.0"
 gem "rspec", "~> 3.0"
 

data/Rakefile

@@ -12,13 +12,13 @@ RuboCop::RakeTask.new
 task default: :spec
 
 namespace :spec do
-  desc "Download latest JIS index fixture from relaton-data-jis"
+  desc "Download latest JIS index fixture (index-v2) from relaton-data-jis"
   task :update_index do
     require "net/http"
     require "uri"
 
-    url = "https://raw.githubusercontent.com/relaton/relaton-data-jis/data-v2/index-v1.zip"
-    dest = File.join(__dir__, "spec", "fixtures", "index-v1.zip")
+    url = "https://raw.githubusercontent.com/relaton/relaton-data-jis/v2/index-v2.zip"
+    dest = File.join(__dir__, "spec", "fixtures", "index-v2.zip")
 
     puts "Downloading \#{url} ..."
     uri = URI.parse(url)

data/lib/relaton/jis/bibliography.rb

@@ -6,21 +6,26 @@ module Relaton
       extend self
 
       #
-      # Search JIS by keyword
+      # Search JIS by reference, returning the full candidate collection.
+      #
+      # The reference is parsed into a {Pubid::Jis::Identifier} and matched
+      # against the pubid-based `index-v2` via {HitCollection}. Candidates share
+      # the reference's series and number (a supplement is filed under its base
+      # number), so an edition and its amendments are returned together.
       #
       # @param [String] code JIS document code
       # @param [String, nil] year JIS document year
       #
-      # @return [Relaton::Jis::HitCollection] search result
+      # @return [Relaton::Jis::HitCollection, nil] search result, or nil when
+      #   the reference cannot be parsed
       #
       def search(code, year = nil)
-        index = Relaton::Index.find_or_create(
-          :jis,
-          url: "#{HitCollection::GH_URL}#{INDEXFILE}.zip",
-          file: "#{INDEXFILE}.yaml",
-        )
-        result = index.search(code).sort_by { |h| h[:id] }
-        HitCollection.new code, year, result: result
+        pubid = ::Pubid::Jis::Identifier.parse code
+        pubid.year ||= year.to_i if year
+        HitCollection.new pubid
+      rescue StandardError => e
+        Util.warn "Unable to parse `#{code}` with pubid: #{e.message}"
+        nil
       end
 
       #
@@ -53,13 +58,13 @@ module Relaton
       #
       # Log hint message
       #
-      # @param [Array] result search result
+      # @param [Array] result search result (missed edition years)
       # @param [String] ref reference to search
       # @param [String, nil] year year to search
       #
       def hint(result, ref, year)
         Util.info "Not found.", key: ref
-        if result.any?
+        if result&.any?
           Util.info "TIP: No match for edition year `#{year}`, but " \
                     "matches exist for `#{result.uniq.join('`, `')}`.", key: ref
         end

data/lib/relaton/jis/data_fetcher.rb

@@ -23,6 +23,26 @@ module Relaton
         @index ||= Relaton::Index.find_or_create :jis, file: "#{INDEXFILE}.yaml"
       end
 
+      # Pubid-based index built in parallel with the legacy string index. The
+      # pool keys by type, so requesting a second :jis index with a different
+      # file evicts the v1 Type from the pool, but we keep our own reference in
+      # @index, so both stay live for the duration of the crawl.
+      def index_v2
+        @index_v2 ||= Relaton::Index.find_or_create(
+          :jis, file: "#{INDEXFILE_V2}.yaml", pubid_class: ::Pubid::Jis::Identifier
+        )
+      end
+
+      # Parse a primary docidentifier string into a pubid identifier; nil (with
+      # a warning) if pubid can't parse it, so a single bad id never aborts the
+      # crawl or corrupts index-v2.
+      def pubid(id)
+        ::Pubid::Jis::Identifier.parse id
+      rescue StandardError => e
+        Util.warn "Failed to parse `#{id}` with pubid: #{e.message}"
+        nil
+      end
+
       def to_yaml(bib)
         Item.to_yaml bib
       end
@@ -69,6 +89,7 @@ module Relaton
         resp = agent.get "#{URL}W11M0070/index"
         parse_page resp
         index.save
+        index_v2.save
         report_errors
       end
 
@@ -150,6 +171,8 @@ module Relaton
             @files << file
             File.write file, serialize(bib), encoding: "UTF-8"
             index.add_or_update id, file
+            pid = pubid id
+            index_v2.add_or_update pid, file if pid
           end
         end
       end

data/lib/relaton/jis/hit.rb

@@ -16,27 +16,45 @@ module Relaton
       end
 
       #
-      # Check if hit matches reference
+      # Check if the hit matches the collection's reference.
       #
-      # @param [Hash] ref_parts parts of reference
-      # @param [String, nil] year year
-      # @param [Boolean] all_parts check all parts of reference
+      # The candidate must be the same document type (so a plain standard query
+      # never matches its amendments) and share series and number. Part is
+      # compared only when the reference names a specific part and `all_parts`
+      # is off. Year is filtered separately by {HitCollection}.
       #
-      # @return [Boolean] true if hit matches reference
+      # @param [Boolean] all_parts match any part of the document
       #
-      def eq?(ref_parts, year = nil, all_parts: false) # rubocop:disable Metrics/CyclomaticComplexity, Metrics/AbcSize, Metrics/PerceivedComplexity
-        part_match?(ref_parts, all_parts) &&
-          (year.nil? || year == id_parts[:year]) &&
-          expl_match?(ref_parts) && amd_match?(ref_parts)
+      # @return [Boolean] true if the hit matches
+      #
+      def matches?(all_parts: false)
+        cand = pubid
+        return false unless cand && same_document?(cand)
+        return true if all_parts || reference_partless?
+
+        Array(cand.parts).map(&:to_s) == reference_parts
       end
 
       #
-      # Return parts of document id
+      # The hit's pubid identifier. `index-v2` rows are already deserialized to
+      # {Pubid::Jis::Identifier} via `pubid_class`; a Hash or String id is
+      # converted for robustness.
       #
-      # @return [Hash] hash with parts of document id
+      # @return [Pubid::Jis::Identifier, nil] identifier, or nil when it cannot
+      #   be built
       #
-      def id_parts
-        @id_parts ||= hit_collection.parse_ref hit[:id]
+      def pubid
+        return @pubid if defined? @pubid
+
+        id = hit[:id]
+        @pubid = case id
+                 when Hash then ::Pubid::Jis::Identifier.from_hash id
+                 when String then ::Pubid::Jis::Identifier.parse id
+                 else id
+                 end
+      rescue StandardError
+        Util.warn "Unable to create an identifier from `#{hit[:id]}`"
+        @pubid = nil
       end
 
       # @return [Relaton::Jis::Item]
@@ -52,24 +70,23 @@ module Relaton
 
       private
 
-      def part_match?(ref_parts, all_parts)
-        id_parts[:code].include?(ref_parts[:code]) &&
-          (all_parts || ref_parts[:part].nil? ||
-           ref_parts[:part] == id_parts[:part])
+      def reference
+        hit_collection.pubid
+      end
+
+      # Same document type, series and number as the reference.
+      def same_document?(cand)
+        cand.instance_of?(reference.class) &&
+          cand.series == reference.series &&
+          cand.number.to_s == reference.number.to_s
       end
 
-      def expl_match?(ref_parts)
-        (ref_parts[:expl].nil? || !id_parts[:expl].nil?) &&
-          (ref_parts[:expl_num].nil? ||
-           ref_parts[:expl_num] == id_parts[:expl_num])
+      def reference_partless?
+        reference.parts.nil? || reference.parts.empty?
       end
 
-      def amd_match?(ref_parts) # rubocop:disable Metrics/AbcSize
-        (ref_parts[:amd].nil? || !id_parts[:amd].nil?) &&
-          (ref_parts[:amd_num].nil? ||
-           ref_parts[:amd_num] == id_parts[:amd_num]) &&
-          (ref_parts[:amd_year].nil? ||
-           ref_parts[:amd_year] == id_parts[:amd_year])
+      def reference_parts
+        Array(reference.parts).map(&:to_s)
       end
     end
   end

data/lib/relaton/jis/hit_collection.rb

@@ -5,29 +5,39 @@ require_relative "hit"
 module Relaton
   module Jis
     class HitCollection < Core::HitCollection
-      GH_URL = "https://raw.githubusercontent.com/relaton/relaton-data-jis/data-v2/"
+      GH_URL = "https://raw.githubusercontent.com/relaton/relaton-data-jis/v2/"
 
       #
-      # Initialize hit collection
+      # Initialize hit collection.
       #
-      # @param [String] text reference
-      # @param [String, nil] year year
-      # @param [Array<Hash>] result search results
+      # Searches the pubid-based `index-v2` for every entry sharing the
+      # reference's series and number (a supplement is filed under its base
+      # number, so editions and amendments come back together). Narrowing to a
+      # specific type/year/part happens later in {#find}.
       #
-      def initialize(text, year = nil, result:)
-        super text, year
-        @array = result.map { |h| Hit.create h, self }
+      # @param [Pubid::Jis::Identifier] pubid parsed reference
+      #
+      def initialize(pubid)
+        super(pubid, pubid.year&.to_s)
+        @array = index.search(pubid) { |row| same_base? row[:id] }
+          .map { |row| Hit.new row, self }
+          .sort_by { |hit| hit.pubid.to_s }
+      end
+
+      # @return [Pubid::Jis::Identifier] parsed reference
+      def pubid
+        ref
       end
 
       #
-      # Find hit in collection
+      # Find the best hit for the reference.
       #
-      # @return [Relaton::Bib::ItemData, Array<String>]
+      # @return [Relaton::Bib::ItemData, Array<Integer>] the matching item, or
+      #   the list of available edition years when none match the requested year
       #
       def find
-        ref_year = year || ref_parts[:year]
-        if ref_year
-          find_by_year ref_year
+        if pubid.year
+          find_by_year pubid.year
         else
           find_all_years
         end
@@ -36,75 +46,81 @@ module Relaton
       def find_by_year(ref_year)
         missed_years = []
         @array.each do |hit|
-          return hit.item if hit.eq? ref_parts, ref_year
+          next unless hit.matches?
 
-          missed_years << hit.id_parts[:year] if hit.eq?(ref_parts)
+          return hit.item if hit.pubid.year.to_s == ref_year.to_s
+
+          missed_years << hit.pubid.year
         end
         missed_years
       end
 
-      def find_all_years # rubocop:disable Metrics/AbcSize
-        hits = @array.select { |hit| hit.eq? ref_parts }
-        return [] if hits.empty?
+      # The main item is the latest edition of the requested type; every other
+      # candidate sharing the series and number (older editions, amendments,
+      # explanations) is attached as an `instanceOf` relation.
+      def find_all_years
+        editions = @array.select(&:matches?)
+        return [] if editions.empty?
 
-        item = hits.max_by { |i| i.id_parts[:year].to_i }.item
-        item_id = item.docidentifier.first.content
-        parent = item.to_most_recent_reference
-        hits.each do |hit|
-          next if hit.hit[:id] == item_id
+        item = editions.max_by { |hit| hit.pubid.year.to_i }.item
+        attach_relations item.to_most_recent_reference,
+                         item.docidentifier.first.content
+      end
 
-          parent.relation << create_relation(hit)
-        end
-        parent
+      # The lowest-numbered part becomes the all-parts umbrella; every candidate
+      # sharing the series and number is attached as an `instanceOf` relation.
+      def find_all_parts
+        parts = @array.select { |hit| hit.matches? all_parts: true }
+        lowest = parts.min_by { |hit| hit.pubid.parts.first.to_i }
+        item = lowest.item.to_all_parts
+        attach_relations item, item.docidentifier.first.content
       end
 
-      def find_all_parts # rubocop:disable Metrics/AbcSize
-        hits = @array.select { |hit| hit.eq? ref_parts, all_parts: true }
-        item = hits.min_by { |i| i.id_parts[:part].to_i }.item.to_all_parts
-        hits.each do |hit|
-          next if hit.hit[:id] == item.docidentifier.first.content
+      # Attach every candidate except `skip_id` to `umbrella` as an `instanceOf`
+      # relation and return the umbrella.
+      def attach_relations(umbrella, skip_id)
+        @array.each do |hit|
+          next if hit.pubid.to_s == skip_id
 
-          item.relation << create_relation(hit)
+          umbrella.relation << create_relation(hit)
         end
-        item
+        umbrella
       end
 
       def create_relation(hit)
-        docid = Docidentifier.new(
-          content: hit.hit[:id], type: "JIS", primary: true,
+        id = hit.pubid.to_s
+        docid = Docidentifier.new content: id, type: "JIS", primary: true
+        bibitem = ItemData.new(
+          formattedref: Bib::Formattedref.new(content: id),
+          docidentifier: [docid],
         )
-        bibitem = Bib::ItemData.new(
-          formattedref: Bib::Formattedref.new(content: hit.hit[:id]), docidentifier: [docid],
+        Relation.new type: "instanceOf", bibitem: bibitem
+      end
+
+      # Index of pubid identifiers (`index-v2`), deserialized via `pubid_class`.
+      def index
+        @index ||= Relaton::Index.find_or_create(
+          :jis,
+          url: "#{GH_URL}#{INDEXFILE_V2}.zip",
+          file: "#{INDEXFILE_V2}.yaml",
+          pubid_class: ::Pubid::Jis::Identifier,
         )
-        Bib::Relation.new(type: "instanceOf", bibitem: bibitem)
       end
 
-      #
-      # Return parts of reference
-      #
-      # @return [Hash] hash with parts of reference
-      #
-      def ref_parts
-        @ref_parts ||= parse_ref ref
+      private
+
+      # Broad candidate filter: same series and number as the reference. For a
+      # supplement (amendment/corrigendum/explanation) the document series and
+      # number live on `base`, so compare against that.
+      def same_base?(candidate)
+        cand = base_of candidate
+        ours = base_of pubid
+        cand.series == ours.series && cand.number.to_s == ours.number.to_s
       end
 
-      #
-      # Parse reference
-      #
-      # @param [String] ref reference
-      #
-      # @return [Hash] hash with parts of reference
-      #
-      def parse_ref(ref)
-        %r{
-          ^(?<code>\w+\s\w\s?\w+)
-          (?:-(?<part>\w+))?
-          (?::(?<year>\d{4}))?
-          (?:/(?<expl>EXPL(?:ANATION)?)(?:\s(?<expl_num>\d+))?)?
-          (?:/(?<amd>AMDENDMENT)(?:\s(?<amd_num>\d+)(?::(?<amd_year>\d{4}))?)?)?
-        }x =~ ref
-        { code: code, part: part, year: year, expl: expl, expl_num: expl_num,
-          amd: amd, amd_num: amd_num, amd_year: amd_year }
+      def base_of(identifier)
+        supplement = identifier.respond_to?(:base) && identifier.base
+        supplement || identifier
       end
     end
   end

data/lib/relaton/jis/processor.rb

@@ -47,7 +47,7 @@ module Relaton
       def remove_index_file
         require_relative "../jis"
         Relaton::Index.find_or_create(
-          :jis, url: true, file: "#{INDEXFILE}.yaml"
+          :jis, url: true, file: "#{INDEXFILE_V2}.yaml"
         ).remove_file
       end
     end

data/lib/relaton/jis/scraper.rb

@@ -28,7 +28,7 @@ module Relaton
         contributors << eg_contributor if eg_contributor
         attrs = ATTRS.to_h { |attr| [attr, send("fetch_#{attr}")] }
         attrs[:contributor] = contributors
-        Bib::ItemData.new(**attrs)
+        ItemData.new(**attrs)
       end
 
       def fetch_title
@@ -213,7 +213,7 @@ module Relaton
       end
 
       def fetch_structuredidentifier
-        Iso::StructuredIdentifier.new(
+        StructuredIdentifier.new(
           project_number: Iso::ProjectNumber.new(content: fetch_docnumber),
           type: "JIS",
         )

data/lib/relaton/jis/version.rb

@@ -2,6 +2,6 @@
 
 module Relaton
   module Jis
-    VERSION = "2.1.0"
+    VERSION = "2.2.0.pre.alpha.1"
   end
 end

data/lib/relaton/jis.rb

@@ -2,6 +2,7 @@
 
 require "net/http"
 require "mechanize"
+require "pubid"
 require "relaton/iso"
 require "relaton/index"
 require_relative "jis/version"
@@ -15,6 +16,8 @@ require_relative "jis/bibliography"
 module Relaton
   module Jis
     INDEXFILE = "index-v1"
+    # Pubid-based index produced alongside index-v1 during the pubid migration.
+    INDEXFILE_V2 = "index-v2"
 
     class Error < StandardError; end
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: relaton-jis
 version: !ruby/object:Gem::Version
-  version: 2.1.0
+  version: 2.2.0.pre.alpha.1
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-05-04 00:00:00.000000000 Z
+date: 2026-06-26 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: mechanize
@@ -24,48 +24,62 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '2.10'
+- !ruby/object:Gem::Dependency
+  name: pubid
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 2.0.0.pre.alpha.3
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 2.0.0.pre.alpha.3
 - !ruby/object:Gem::Dependency
   name: relaton-core
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.0.13
+        version: 2.2.0.pre.alpha.1
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.0.13
+        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: 0.2.0
+        version: 2.2.0.pre.alpha.1
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: 0.2.0
+        version: 2.2.0.pre.alpha.1
 - !ruby/object:Gem::Dependency
   name: relaton-iso
   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
 description: 'Relaton::Jis: retrieve IETF Standards for bibliographic use using the
   BibliographicItem model'
 email:
@@ -75,17 +89,11 @@ extensions: []
 extra_rdoc_files: []
 files:
 - ".rspec"
-- ".rubocop.yml"
 - CLAUDE.md
 - Gemfile
 - LICENSE.txt
 - README.adoc
 - Rakefile
-- grammars/basicdoc.rng
-- grammars/biblio-standoc.rng
-- grammars/biblio.rng
-- grammars/relaton-jis-compile.rng
-- grammars/relaton-jis.rng
 - lib/relaton/jis.rb
 - lib/relaton/jis/bibdata.rb
 - lib/relaton/jis/bibitem.rb
@@ -105,7 +113,6 @@ files:
 - lib/relaton/jis/structured_identifier.rb
 - lib/relaton/jis/util.rb
 - lib/relaton/jis/version.rb
-- relaton-jis.gemspec
 - sig/relaton_jis.rbs
 homepage: https://github.com/relaton/relaton-jis
 licenses:
@@ -120,7 +127,7 @@ 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