relaton 3.0.0.pre.alpha.4 → 3.0.0.pre.alpha.6

data/lib/relaton/ecma/bibliography.rb

@@ -43,7 +43,8 @@ module Relaton
         # `ECMA-418` does not match `ECMA-418-1`. The class must be identical,
         # which keeps `ECMA-100` and `ECMA TR/100` apart.
         #
-        # @param ref [String] the ECMA reference (e.g. "ECMA-6", "ECMA-269 ed3 vol2")
+        # @param ref [String, Pubid::Ecma::Identifier] the ECMA reference
+        #   (e.g. "ECMA-6", "ECMA-269 ed3 vol2"), or its parse
         #
         # @return [Array<Hash>] matching index rows
         #
@@ -62,7 +63,7 @@ module Relaton
         # regex accepted, so every reference shape the flavor has to handle
         # parses without normalization here.
         #
-        # @param ref [String]
+        # @param ref [String, Pubid::Ecma::Identifier]
         # @return [Pubid::Ecma::Identifier, nil]
         #
         # An unrecognized reference **raises**; like ISO, ETSI and 3GPP we let it
@@ -73,20 +74,24 @@ module Relaton
         # Rescuing here would collapse "this identifier is malformed" into "no
         # such document", leaving a caller unable to tell them apart.
         def parse_ref(ref)
+          # A pubid from Relaton::Db (relaton#205) is used as it is.
+          return ref unless ref.is_a?(String)
+
           ::Pubid::Ecma::Identifier.parse ref.to_s.strip
         end
 
-        # @param code [String] the ECMA standard Code to look up (e..g "ECMA-6")
+        # @param ref [String, Pubid::Ecma::Identifier] the ECMA reference
+        #   (e.g. "ECMA-6"), or its parse from Relaton::Db (relaton#205)
         # @param year [String] not used
         # @param opts [Hash] not used
         # @return [Relaton::Ecma::ItemData] Relaton of reference
-        def get(code, _year = nil, _opts = {})
-          Util.info "Fetching from Relaton repository ...", key: code
-          result = fetch_doc(code)
+        def get(ref, _year = nil, _opts = {})
+          Util.info "Fetching from Relaton repository ...", key: ref.to_s
+          result = fetch_doc(ref)
           if result
-            Util.info "Found: `#{result.docidentifier.first.content}`", key: code
+            Util.info "Found: `#{result.docidentifier.first.content}`", key: ref.to_s
           else
-            Util.info "Not found.", key: code
+            Util.info "Not found.", key: ref.to_s
           end
           result
         end
@@ -106,7 +111,7 @@ module Relaton
         #
         # `r[:file]` breaks the tie, because the index sort is not stable.
         #
-        # @param ref [String]
+        # @param ref [String, Pubid::Ecma::Identifier]
         # @return [Hash, nil]
         #
         def best_match(ref)
@@ -127,8 +132,8 @@ module Relaton
           edition.to_s.split(".").map(&:to_i)
         end
 
-        def fetch_doc(code)
-          row = best_match code
+        def fetch_doc(ref)
+          row = best_match ref
           return unless row
 
           url = "#{ENDPOINT}#{row[:file]}"
@@ -137,11 +142,11 @@ module Relaton
         rescue Mechanize::ResponseCodeError => e
           return if e.response_code == "404"
 
-          raise Relaton::RequestError, "No document found for #{code} reference. #{e.message}"
+          raise Relaton::RequestError, "No document found for #{ref} reference. #{e.message}"
         rescue Mechanize::RedirectLimitReachedError, Timeout::Error,
             Mechanize::UnauthorizedError, Mechanize::UnsupportedSchemeError,
             Mechanize::ResponseReadError, Mechanize::ChunkedTerminationError => e
-          raise Relaton::RequestError, "No document found for #{code} reference. #{e.message}"
+          raise Relaton::RequestError, "No document found for #{ref} reference. #{e.message}"
         end
       end
     end

data/lib/relaton/ecma/data_fetcher.rb

@@ -128,7 +128,7 @@ module Relaton
       def to_yaml(bib) = bib.to_yaml
       def to_bibxml(bib) = bib.to_rfcxml
 
-      # @param hit [Nokogiri::XML::Element]
+      # @param hit [Moxml::Element]
       def parse_page(hit) # rubocop:disable Metrics/AbcSize, Metrics/MethodLength
         DataParser.new(hit, @errors).parse.each { |item| write_file item }
       end

data/lib/relaton/ecma/data_parser.rb

@@ -6,7 +6,7 @@ module Relaton
       #
       # Initialize parser
       #
-      # @param [Nokogiri::XML::Element] hit document hit
+      # @param [Moxml::Element] hit document hit
       # @param [Hash] errors error tracking hash
       #
       def initialize(hit, errors = {})

data/lib/relaton/ecma/edition_parser.rb

@@ -21,7 +21,7 @@ module Relaton
         docid = @bib[:docidentifier]
         @doc.xpath('//div[@id="main"]/div[1]/div/main/article/div/div/standard/div/ul/li').map do |hit|
           bib = @bib.dup
-          id, ed, bib[:date], vol = edition_id_parts hit.at("./span", "./a").text
+          id, ed, bib[:date], vol = edition_id_parts hit.at_xpath("./span|./a").text
           bib[:source] = edition_source(hit) + edition_translation_source(ed)
           next if ed.nil? || ed.empty?
 
@@ -56,7 +56,7 @@ module Relaton
       end
 
       def edition_source(hit)
-        es = { "src" => hit.at("./a"), "pdf" => hit.at("./span/a") }.map do |type, a|
+        es = { "src" => hit.at_xpath("./a"), "pdf" => hit.at_xpath("./span/a") }.map do |type, a|
           Bib::Uri.new(type: type, content: a[:href]) if a
         end.compact
         @errors[:edition_source] &&= es.empty?

data/lib/relaton/ecma/memento_parser.rb

@@ -5,7 +5,7 @@ module Relaton
 
       ATTRS = %i[docidentifier title date source ext].freeze
 
-      # @param [Nokogiri::XML::Element] hit document hit
+      # @param [Moxml::Element] hit document hit
       # @param [Hash] errors error tracking hash
       def initialize(hit:, errors: {})
         @hit = hit
@@ -23,7 +23,7 @@ module Relaton
 
       # @return [Array<Relaton::Ecma::Docidentifier>]
       def fetch_docidentifier
-        code = "ECMA MEM/#{@hit.at('div[1]//p').text}"
+        code = "ECMA MEM/#{@hit.at_xpath('div[1]//p').text}"
         docid = super(code)
         @errors[:memento_docidentifier] &&= docid.empty?
         docid
@@ -31,7 +31,7 @@ module Relaton
 
       # @return [Array<Relaton::Bib::Title>]
       def fetch_title
-        year = @hit.at("div[1]//p").text
+        year = @hit.at_xpath("div[1]//p").text
         content = "\"Memento #{year}\" for year #{year}"
         result = [Bib::Title.new(content: content, language: "en", script: "Latn")]
         @errors[:memento_title] &&= result.empty?
@@ -40,7 +40,7 @@ module Relaton
 
       # @return [Array<Relaton::Bib::Date>]
       def fetch_date
-        date = @hit.at("div[2]//p").text
+        date = @hit.at_xpath("div[2]//p").text
         on = Date.strptime(date, "%B %Y").strftime "%Y-%m"
         result = [Bib::Date.new(type: "published", at: on)]
         @errors[:memento_date] &&= result.empty?

data/lib/relaton/ecma/standard_parser.rb

@@ -5,7 +5,7 @@ module Relaton
 
       ATTRS = %i[docidentifier title date source abstract relation edition ext].freeze
 
-      # @param [Nokogiri::XML::Element] hit document hit
+      # @param [Moxml::Element] hit document hit
       # @param [Mechanize::Page] doc fetched document page
       # @param [Hash] errors error tracking hash
       def initialize(hit:, doc:, errors: {})
@@ -68,7 +68,7 @@ module Relaton
       def fetch_source # rubocop:disable Metrics/AbcSize
         source = []
         source << Bib::Uri.new(type: "src", content: @hit[:href]) if @hit[:href]
-        ref = @doc.at('//div[@class="ecma-item-content-wrapper"]/span/a',
+        ref = @doc.at_xpath('//div[@class="ecma-item-content-wrapper"]/span/a',
                       '//div[@class="ecma-item-content-wrapper"]/a')
         source << Bib::Uri.new(type: "pdf", content: ref[:href]) if ref
         result = source + edition_translation_source(fetch_edition_content)
@@ -80,7 +80,7 @@ module Relaton
       def fetch_relation # rubocop:disable Metrics/AbcSize, Metrics/MethodLength, Metrics/CyclomaticComplexity
         edition_parser = EditionParser.new(doc: @doc, bib: {}, errors: @errors)
         result = @doc.xpath("//ul[@class='ecma-item-archives']/li").filter_map do |rel|
-          ref, ed, date, vol = edition_parser.edition_id_parts rel.at("span").text
+          ref, ed, date, vol = edition_parser.edition_id_parts rel.at_xpath("span").text
           next if ed.nil? || ed.empty?
 
           docid = Docidentifier.new(type: "ECMA", content: ref, primary: true)
@@ -109,7 +109,7 @@ module Relaton
       private
 
       def fetch_edition_content
-        @doc.at('//p[@class="ecma-item-edition"]')&.text&.match(/^\d+(?=(?:st|nd|th|rd))/)&.to_s
+        @doc.at_xpath('//p[@class="ecma-item-edition"]')&.text&.match(/^\d+(?=(?:st|nd|th|rd))/)&.to_s
       end
 
       def edition_translation_source(edition)
@@ -120,8 +120,8 @@ module Relaton
         return [] unless @doc
 
         @doc.xpath("//h2[.='Translations']/following-sibling::ul/li").map do |l|
-          a = l.at("span/a")
-          id = l.at("span").text
+          a = l.at_xpath("span/a")
+          id = l.at_xpath("span").text
           %r{\w+[\d-]+,\s(?<lang>\w+)\sversion,\s(?<ed>[\d.]+)(?:st|nd|rd|th)\sedition} =~ id
           case lang
           when "Japanese"

data/lib/relaton/etsi/bibliography.rb

@@ -6,14 +6,15 @@ module Relaton
     module Bibliography
       SOURCE = "https://raw.githubusercontent.com/relaton/relaton-data-etsi/refs/heads/v2/"
 
-      # @param text [String]
+      # @param ref [String, ::Pubid::Etsi::Identifier]
       # @return [Relaton::Etsi::ItemData, nil]
-      def search(text) # rubocop:disable Metrics/MethodLength, Metrics/AbcSize
+      def search(ref) # rubocop:disable Metrics/MethodLength, Metrics/AbcSize
         # An unrecognized reference raises Pubid::Errors::ParseError; like
         # ISO we let it propagate — the CLI turns it into a friendly message
         # and API callers rescue it themselves. Valid partial refs parse with
         # the omitted refinements (version/date/part) left blank.
-        pubid = ::Pubid::Etsi.parse text
+        # A parsed pubid comes from Relaton::Db (relaton#205); it is only read.
+        pubid = ref.is_a?(String) ? ::Pubid::Etsi.parse(ref) : ref
 
         index = Relaton::Index.find_or_create :etsi, url: "#{SOURCE}#{INDEXFILE}.zip", file: "#{INDEXFILE}.yaml",
                                                      pubid_class: ::Pubid::Etsi::Identifier
@@ -89,19 +90,19 @@ module Relaton
         [(id.version&.version).to_s.split(".").map(&:to_i), id.date.to_s]
       end
 
-      # @param ref [String] the ETSI standard Code to look up
+      # @param ref [String, ::Pubid::Etsi::Identifier] the ETSI standard Code to look up
       # @param year [String, nil] year
       # @param opts [Hash] options
       # @return [Relaton::Etsi::ItemData, nil]
       def get(ref, _year = nil, _opts = {})
-        Util.info "Fetching from Relaton repository ...", key: ref
+        Util.info "Fetching from Relaton repository ...", key: ref.to_s
         result = search(ref)
         unless result
-          Util.info "Not found.", key: ref
+          Util.info "Not found.", key: ref.to_s
           return
         end
 
-        Util.info "Found: `#{result.docidentifier[0].content}`", key: ref
+        Util.info "Found: `#{result.docidentifier[0].content}`", key: ref.to_s
         result
       end
 

data/lib/relaton/etsi/processor.rb

@@ -8,6 +8,7 @@ module Relaton
       def initialize # rubocop:disable Lint/MissingSuper
         @short = :relaton_etsi
         @prefix = "ETSI"
+        @pubid_identifier = :Etsi # Db cache key
         @defaultprefix = %r{^ETSI\s}
         @idtype = "ETSI"
         @datasets = %w[etsi-csv]

data/lib/relaton/gb/bibliography.rb

@@ -9,15 +9,15 @@ module Relaton
     class Bibliography
       class << self
         # rubocop:disable Metrics/MethodLength
-        # @param text [Strin] code of standard for search
+        # @param ref [Strin] code of standard for search
         # @return [RelatonGb::HitCollection]
-        def search(text)
-          case text
+        def search(ref)
+          case ref
           when /^(GB|GJ|GS)/
             # Scrape national standards.
-            Util.info "Fetching from openstd.samr.gov.cn ...", key: text
+            Util.info "Fetching from openstd.samr.gov.cn ...", key: ref
             require_relative "gb_scraper"
-            GbScraper.scrape_page text
+            GbScraper.scrape_page ref
           # when /^ZB/
             # Scrape proffesional.
           # when /^DB/
@@ -26,26 +26,34 @@ module Relaton
             # Enterprise standard
           when %r{^T/[^\s]{2,6}\s}
             # Scrape social standard.
-            Util.info "Fetching from www.ttbz.org.cn ...", key: text
+            Util.info "Fetching from www.ttbz.org.cn ...", key: ref
             require_relative "t_scraper"
-            TScraper.scrape_page text
+            TScraper.scrape_page ref
           else
             # Scrape sector standard.
             require "relaton/gb/sec_scraper"
-            SecScraper.scrape_page text
+            SecScraper.scrape_page ref
           end
         end
         # rubocop:enable Metrics/MethodLength
 
-        # @param code [String] the GB standard Code to look up (e..g "GB/T 20223")
+        # @param ref [String, Pubid::Gb::Identifier] the GB standard Code to
+        #   look up (e.g. "GB/T 20223"), or its parse from Relaton::Db
+        #   (relaton#205)
         # @param year [String] the year the standard was published (optional)
         # @param opts [Hash] options; restricted to :all_parts if all-parts reference is required
         # @return [Relaton::Gb::ItemData, nil]
-        # @raise [Pubid::Errors::ParseError] when the code is not a GB
+        # @raise [Pubid::Errors::ParseError] when the reference is not a GB
         #   identifier
-        def get(code, year = nil, opts = {})
+        def get(ref, year = nil, opts = {}) # rubocop:disable Metrics/AbcSize,Metrics/CyclomaticComplexity,Metrics/PerceivedComplexity
           require "pubid"
-          pubid = ::Pubid::Gb::Identifier.parse(code)
+          # A parsed pubid comes from Relaton::Db (relaton#205); it is never
+          # mutated (`exclude` below copies).
+          pubid = ref.is_a?(String) ? ::Pubid::Gb::Identifier.parse(ref) : ref
+          if pubid.all_parts?
+            opts = opts.merge(all_parts: true)
+            pubid = pubid.identifiers.first
+          end
           year = (year || pubid.year)&.to_s
           pubid = pubid.exclude(:year)
           pubid.part = "1" if opts[:all_parts]

data/lib/relaton/gb/gb_scraper.rb

@@ -1,7 +1,7 @@
 # encoding: UTF-8
 # frozen_string_literal: true
 
-require "nokogiri"
+require "moxml"
 require_relative "scraper"
 
 module Relaton
@@ -20,10 +20,10 @@ module Relaton
           hits = doc.xpath(
             "//table[contains(@class, 'result_list')]/tbody[2]/tr",
           ).map do |h|
-            ref = h.at "./td[2]/a"
+            ref = h.at_xpath "./td[2]/a"
             pid = ref[:onclick].match(/[0-9A-F]+/).to_s
-            status = h.at("./td[7]").text.strip
-            rdate = h.at("./td[8]").text.strip
+            status = h.at_xpath("./td[7]").text.strip
+            rdate = h.at_xpath("./td[8]").text.strip
             Hit.new pid: pid, docref: ref.text, scraper: self,
                     release_date: rdate, status: status
           end
@@ -52,7 +52,7 @@ module Relaton
         #   * :type [String]
         #   * :name [String]
         # def get_committee(doc, _ref)
-        #   name = doc.at("//div[contains(., '归口单位') or contains(., '归口部门')]/following-sibling::div")
+        #   name = doc.at_xpath("//div[contains(., '归口单位') or contains(., '归口部门')]/following-sibling::div")
         #   Committee.new(type: "technical", content: name.text.delete("\r\n\t\t"))
         # end
       end

data/lib/relaton/gb/scraper.rb

@@ -15,7 +15,7 @@ module Relaton
 
       @prefixes = nil
 
-      # @param doc [Nokogiri::HTML::Document]
+      # @param doc [Moxml::Document]
       # @param src [String]
       # @param hit [RelatonGb::Hit]
       # @return [Hash]
@@ -41,7 +41,7 @@ module Relaton
         [Docidentifier.new(content: docref, type: "Chinese Standard", primary: true)]
       end
 
-      # @param doc [Nokogiri::HTML::Document]
+      # @param doc [Moxml::Document]
       # @param docref [Strings]
       # @return [Array<Relaton::Bib::Contributor>]
       def get_contributors(doc, docref)
@@ -67,22 +67,22 @@ module Relaton
         Bib::TypedLocalizedString.new language: lang, content: content
       end
 
-      # @param doc [Nokogiri::HTML::Document]
+      # @param doc [Moxml::Document]
       # @return [Array<Relaton::Bib::Title>]
       def get_titles(doc)
-        tzh = doc.at("//td[contains(text(), '中文标准名称')]/b").text
+        tzh = doc.at_xpath("//td[contains(text(), '中文标准名称')]/b").text
         titles = Relaton::Bib::Title.from_string tzh, "zh", "Hans"
-        ten = doc.at("//td[contains(text(), '英文标准名称')]").text.match(/[\w\s]+/).to_s
+        ten = doc.at_xpath("//td[contains(text(), '英文标准名称')]").text.match(/[\w\s]+/).to_s
         return titles if ten.empty?
 
         titles + Relaton::Bib::Title.from_string(ten, "en", "Latn")
       end
 
-      # @param doc [Nokogiri::HTML::Document]
+      # @param doc [Moxml::Document]
       # @param status [String, NilClass]
       # @return [Relaton::Bib::Status]
       def get_status(doc, status = nil)
-        status ||= doc.at("//td[contains(., '标准状态')]/span")&.text&.strip
+        status ||= doc.at_xpath("//td[contains(., '标准状态')]/span")&.text&.strip
         return unless STAGES[status]
 
         stage = Bib::Status::Stage.new content: STAGES[status]
@@ -91,17 +91,17 @@ module Relaton
 
       private
 
-      # @param doc [Nokogiri::HTML::Document]
+      # @param doc [Moxml::Document]
       # @return [Array<String>]
       def get_ccs(doc)
-        code = doc.at("//div[contains(text(), '中国标准分类号')]/following-sibling::div").text.strip
+        code = doc.at_xpath("//div[contains(text(), '中国标准分类号')]/following-sibling::div").text.strip
         [CCS.new(code: code)]
       end
 
-      # @param doc [Nokogiri::HTML::Document]
+      # @param doc [Moxml::Document]
       # @return [Array<Relaton::Bib::ICS>]
       def get_ics(doc)
-        ics = doc.at("//div[contains(text(), '国际标准分类号')]/following-sibling::div"\
+        ics = doc.at_xpath("//div[contains(text(), '国际标准分类号')]/following-sibling::div"\
                     " | //dt[contains(text(), '国际标准分类号')]/following-sibling::dd")
         return [] unless ics
 
@@ -109,10 +109,10 @@ module Relaton
         [Bib::ICS.new(code: code)]
       end
 
-      # @param doc [Nokogiri::HTML::Document]
+      # @param doc [Moxml::Document]
       # @return [String]
       def get_scope(doc)
-        issued = doc.at("//div[contains(., '发布单位')]/following-sibling::div")
+        issued = doc.at_xpath("//div[contains(., '发布单位')]/following-sibling::div")
         case issued&.text
         when /国家标准/ then "national"
         when /^行业标准/ then "sector"
@@ -150,12 +150,12 @@ module Relaton
         (Bib::Uri.new(type: "src", content: src))
       end
 
-      # @param doc [Nokogiri::HTML::Document]
+      # @param doc [Moxml::Document]
       # @return [Array<Hash>]
       #   * :type [String] type of date
       #   * :on [String] date
       def get_dates(doc)
-        date = doc.at("//div[contains(text(), '发布日期')]/following-sibling::div"\
+        date = doc.at_xpath("//div[contains(text(), '发布日期')]/following-sibling::div"\
                       " | //dt[contains(text(), '发布日期')]/following-sibling::dd")
         [Bib::Date.new(type: "published", at: date.text.delete("\r\n\t\t"))]
       end
@@ -177,7 +177,7 @@ module Relaton
         Doctype.new content: "standard"
       end
 
-      # @param doc [Nokogiri::HTML::Document]
+      # @param doc [Moxml::Document]
       # @param ref [String]
       # @return [Relaton::Gb::GbType]
       def get_gbtype(doc, ref)

data/lib/relaton/gb/sec_scraper.rb

@@ -3,7 +3,7 @@
 
 require "net/http"
 require "json"
-require "nokogiri"
+require "moxml"
 require_relative "scraper"
 require_relative "item"
 require_relative "hit_collection"
@@ -44,7 +44,7 @@ module Relaton
         def scrape_doc(hit)
           src = "https://hbba.sacinfo.org.cn/stdDetail/#{hit.pid}"
           page_uri = URI src
-          doc = Nokogiri::HTML Net::HTTP.get(page_uri)
+          doc = Moxml.new.parse_html Net::HTTP.get(page_uri)
           ItemData.new(**scrapped_data(doc, src, hit))
         rescue SocketError, Timeout::Error, Errno::EINVAL, Errno::ECONNRESET, EOFError,
               Net::HTTPBadResponse, Net::HTTPHeaderSyntaxError, Net::ProtocolError,
@@ -54,14 +54,14 @@ module Relaton
 
         private
 
-        # @param doc [Nokogiri::HTML::Document]
+        # @param doc [Moxml::Document]
         # @return [Array<Relaton::Bib::Title>]
         def get_titles(doc)
-          tzh = doc.at("//h4").text.delete("\r\n\t")
+          tzh = doc.at_xpath("//h4").text.delete("\r\n\t")
           Bib::Title.from_string(tzh, "zh", "Hans")
         end
 
-        # @param _doc [Nokogiri::HTML::Document]
+        # @param _doc [Moxml::Document]
         # @param ref [String]
         # @return [Hash]
         #   * :type [String]
@@ -72,16 +72,16 @@ module Relaton
         #   { type: "technical", name: name }
         # end
 
-        # @param _doc [Nokogiri::HTML::Document]
+        # @param _doc [Moxml::Document]
         # @return [String]
         def get_scope(_doc)
           "sector"
         end
 
-        # @param doc [Nokogiri::HTML::Document]
+        # @param doc [Moxml::Document]
         # @return [Array<String>]
         def get_ccs(doc)
-          array(doc.at("//dt[contains(text(), '中国标准分类号')]/following-sibling::dd")).map do |cc|
+          array(doc.at_xpath("//dt[contains(text(), '中国标准分类号')]/following-sibling::dd")).map do |cc|
             text = Cnccs.fetch(cc.text.strip)&.description
             CCS.new code: cc.text, text: text
           end

data/lib/relaton/gb/t_scraper.rb

@@ -1,7 +1,7 @@
 # encoding: UTF-8
 # frozen_string_literal: true
 
-require "nokogiri"
+require "moxml"
 require_relative "scraper"
 require_relative "hit_collection"
 require_relative "hit"
@@ -23,8 +23,8 @@ module Relaton
           xpath = '//table[contains(@class, "standard_list_table")]/tr/td/a'
           t_xpath = "../preceding-sibling::td[4]"
           hits = doc.xpath(xpath).map do |h|
-            docref = h.at(t_xpath).text.gsub(/â\u0080\u0094/, "-")
-            status = h.at("../preceding-sibling::td[1]").text.delete "\r\n"
+            docref = h.at_xpath(t_xpath).text.gsub(/â\u0080\u0094/, "-")
+            status = h.at_xpath("../preceding-sibling::td[1]").text.delete "\r\n"
             pid = h[:href].sub(%r{/$}, "")
             Hit.new pid: pid, docref: docref, status: status, scraper: self
           end
@@ -55,7 +55,7 @@ module Relaton
         private
 
         # rubocop:disable Metrics/MethodLength
-        # @param doc [Nokogiri::HTML::Document]
+        # @param doc [Moxml::Document]
         # @param src [String]
         # @param hit [RelatonGb::Hit]
         # @return [Hash]
@@ -89,7 +89,7 @@ module Relaton
 
         def get_titles(doc)
           xpz = '//td[contains(.,"中文标题")]/following-sibling::td[1]'
-          titles = Bib::Title.from_string doc.at(xpz)
+          titles = Bib::Title.from_string doc.at_xpath(xpz)
             .text, "zh", "Hans"
           xpe = '//td[contains(.,"英文标题")]/following-sibling::td[1]'
           ten = doc.xpath(xpe).text

data/lib/relaton/gost/processor.rb

@@ -12,6 +12,7 @@ module Relaton
       def initialize
         @short = :relaton_gost
         @prefix = "GOST"
+        @pubid_identifier = :Gost # Db cache key
         # Both Latin "GOST" and Cyrillic "ГОСТ" route here. The trailing
         # \b keeps the prefix from swallowing longer tokens ("GOSTA …").
         @defaultprefix = %r{^(?:GOST|ГОСТ)\b}

data/lib/relaton/iala/bibliography.rb

@@ -8,15 +8,15 @@ module Relaton
       class << self
         # Search for an IALA publication by its identifier.
         #
-        # @param text [String] the IALA reference to look up (e.g. "IALA S1070")
+        # @param ref [String] the IALA reference to look up (e.g. "IALA S1070")
         # @param _year [String, nil] optional edition/year filter
         # @param _opts [Hash] options (unused)
         # @return [Relaton::Iala::Item, nil]
-        def search(text, _year = nil, _opts = {})
-          Util.info "Fetching from Relaton repository ...", key: text
-          row = best_match text
+        def search(ref, _year = nil, _opts = {})
+          Util.info "Fetching from Relaton repository ...", key: ref.to_s
+          row = best_match ref
           unless row
-            Util.info "Not found.", key: text
+            Util.info "Not found.", key: ref.to_s
             return
           end
 
@@ -27,7 +27,7 @@ module Relaton
           end
 
           item = Relaton::Iala::Item.from_yaml resp.body
-          Util.info "Found: `#{item.docidentifier.first&.content}`", key: text
+          Util.info "Found: `#{item.docidentifier.first&.content}`", key: ref.to_s
           item.tap { |i| i.fetched = Date.today.to_s }
         rescue SocketError, Errno::EINVAL, Errno::ECONNRESET, EOFError,
                Net::HTTPBadResponse, Net::HTTPHeaderSyntaxError,
@@ -59,7 +59,7 @@ module Relaton
         # must never match each other. An `Annex` row carries its base
         # identifier, which `===` compares with its own subset match.
         #
-        # @param text [String]
+        # @param ref [String]
         # @return [Hash, nil]
         #
         # The substring-scan fallback this used to take when parsing failed is
@@ -68,8 +68,8 @@ module Relaton
         # ids, so an ambiguous reference silently resolved to whichever row
         # sorted first, instead of telling the caller the reference is not an
         # identifier. ecma, w3c and xsf never had one.
-        def best_match(text)
-          pubid = parse_ref text
+        def best_match(ref)
+          pubid = parse_ref ref
           rows = index.search(pubid)
           rows.max_by { |r| [edition_key(r[:id].edition), language_key(r[:id]), r[:file]] }
         end
@@ -114,7 +114,7 @@ module Relaton
         # zero-pads the number to its type's canonical width, so `IALA M1`,
         # `M0001` and `R1016:ed2.0(F)` all parse without normalization here.
         #
-        # @param text [String]
+        # @param ref [String, Pubid::Iala::Identifier]
         # @return [Pubid::Iala::Identifier, nil]
         #
         # An unrecognized reference **raises**; like ISO, ETSI and 3GPP we let it
@@ -124,8 +124,11 @@ module Relaton
         # logs it through the `StandardError` arm at `lib/relaton/db.rb:122`.
         # Rescuing here would collapse "this identifier is malformed" into "no
         # such document", leaving a caller unable to tell them apart.
-        def parse_ref(text)
-          ::Pubid::Iala::Identifier.parse text.to_s.strip
+        def parse_ref(ref)
+          # A parsed pubid comes from Relaton::Db (relaton#205); it is used as it is.
+          return ref unless ref.is_a?(String)
+
+          ::Pubid::Iala::Identifier.parse ref.to_s.strip
         end
 
         # The index is pubid-backed: `pubid_class:` is what makes

data/lib/relaton/iala/processor.rb

@@ -8,6 +8,7 @@ module Relaton
       def initialize
         @short = :relaton_iala
         @prefix = "IALA"
+        @pubid_identifier = :Iala # Db cache key
         @defaultprefix = %r{^IALA\s}
         @idtype = "IALA"
       end