isodoc 3.5.0 → 3.5.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: c07d19cb743ed84ffc6b1ae61b5b10ddb397cf1aa88781db27dbbd2c3aa63eee
-  data.tar.gz: 6bd29f3cec7aea37250441b3b99f58b36b918641cd0dfc293a030b0ad3cf19c6
+  metadata.gz: ec71cedc759d905b416358b3accb10738478de4ca36ea037173bf615673e38ff
+  data.tar.gz: 6025a91f6810c7f7c901d2c91933000dce2ca94017e0b9978450f725f8e34649
 SHA512:
-  metadata.gz: 72125bdfc8955528ed34b3c456018eecc21aa8cf987ba665b2c13797b02db992fdc9c221e759c623e0aeadf899cde559df337217aa1cb1981ec39b4e1e329d19
-  data.tar.gz: 99236b9f69911c15144794681990944e61771cb238711f21b25d7045e469bb356af1a6098636f2941a1f68d1f64aacd8b31de2a958ca690361493cb1aea4f80f
+  metadata.gz: 807060d18211c7cff066d698336164501fefe85fc0ea649596419ed73ae8228211cee9df19eb90668a635fa06a52e32b7f5af8e20779ea90847cba71764b58fa
+  data.tar.gz: '0128d8924c4274f56b343026f42275f1ffd880ed844873fc7bd74f38b5d68fd1e6781a39a461bd65c19b913dd29a163b74ade0b2176f6f7df5b4da652ee8d3de'

data/isodoc.gemspec

@@ -45,6 +45,8 @@ Gem::Specification.new do |spec|
   spec.add_dependency "uuidtools"
 
   spec.add_development_dependency "benchmark"
+  spec.add_development_dependency "bigdecimal"
+  spec.add_development_dependency "canon", "= 0.2.3"
   spec.add_development_dependency "debug"
   spec.add_development_dependency "equivalent-xml", "~> 0.6"
   spec.add_development_dependency "guard", "~> 2.14"
@@ -55,7 +57,5 @@ Gem::Specification.new do |spec|
   spec.add_development_dependency "sassc-embedded", "~> 1"
   spec.add_development_dependency "simplecov", "~> 0.15"
   spec.add_development_dependency "timecop", "~> 0.9"
-  spec.add_development_dependency "canon" , "= 0.1.3"
-  spec.add_development_dependency "openssl"
   # spec.metadata["rubygems_mfa_required"] = "true"
 end

data/lib/isodoc/convert.rb

@@ -115,10 +115,15 @@ module IsoDoc
       docxml_var_init(docxml)
       convert_i18n_init(docxml)
       metadata_init(@lang, @script, @locale, @i18n)
+      convert_xref_init
+      [docxml, filename, dir]
+    end
+
+    def convert_xref_init
       xref_init(@lang, @script, self, @i18n,
                 { locale: @locale, bibrender: @bibrender })
+      @xrefs.reqt_models = @reqt_models
       @xrefs.klass.doctype = @doctype
-      [docxml, filename, dir]
     end
 
     def convert_i18n_init(docxml)
@@ -126,7 +131,8 @@ module IsoDoc
       i18n_init(@lang, @script, @locale)
       @bibrender ||= bibrenderer
       @reqt_models = requirements_processor
-        .new({ default: "default", lang: @lang, script: @script,
+        .new({ conv: presentation_xml_converter, default: "default",
+               lang: @lang, script: @script,
                locale: @locale, labels: @i18n.get,
                modspecidentifierbase: @modspecidentifierbase })
     end

data/lib/isodoc/function/cleanup.rb

@@ -136,6 +136,22 @@ module IsoDoc
       end
 
       def symbols_cleanup(docxml); end
+
+      # Atomic group (?>...) prevents catastrophic backtracking across `>`
+      # characters. Quoted alternatives listed first so `/` inside
+      # attribute values (e.g. URLs in href="...") is consumed as part
+      # of the quoted string, not rejected by [^>/].
+      def empty_tags(html)
+        void_elements = %w[area base br col embed hr img input link meta
+                           source track wbr]
+        html.gsub(%r{<(\w+)((?>"[^"]*"|'[^']*'|[^>/])*)\s*/>}) do
+          if void_elements.include?($1)
+            $&
+          else
+            "<#{$1}#{$2}></#{$1}>"
+          end
+        end
+      end
     end
   end
 end

data/lib/isodoc/function/utils.rb

@@ -234,7 +234,7 @@ module IsoDoc
         CSV.parse_line(processed, quote_char: "'")
           .each_with_object({}) do |x, acc|
           x.gsub!(COMMA_PLACEHOLDER, ",")
-          m = /^(.+?)=(.+)?$/.match(x) or next
+          m = /^(.+?)=(.*)?$/.match(x) or next
           acc[m[1].to_sym] = m[2].sub(/^(["'])(.+)\1$/, "\\2")
         end
       end

data/lib/isodoc/html_convert.rb

@@ -18,6 +18,14 @@ module IsoDoc
       super
     end
 
+    # Apply empty_tags even when postprocess is skipped (debug mode),
+    # so convert1's output is valid HTML5 (non-void self-closing tags
+    # expanded: <a id="_"/> -> <a id="_"></a>). Idempotent: running
+    # again in postprocess#toHTML is a no-op.
+    def convert1(docxml, filename, dir)
+      empty_tags(super)
+    end
+
     def convert(filename, file = nil, debug = false, output_filename = nil)
       ret = super
       Dir.exist?(tmpimagedir) and Dir["#{tmpimagedir}/*"].empty? and

data/lib/isodoc/html_function/postprocess.rb

@@ -16,7 +16,7 @@ module IsoDoc
       def toHTML(result, filename)
         result = from_xhtml(html_cleanup(to_xhtml(result)))
         result = from_xhtml(move_images(resize_images(to_xhtml(result))))
-        result = html5(script_cdata(inject_script(result)))
+        result = html5(empty_tags(script_cdata(inject_script(result))))
         # Unescape & to & in href attributes after all Nokogiri processing
         result = unescape_amp_in_hrefs(result)
         File.open(filename, "w:UTF-8") { |f| f.write(result) }

data/lib/isodoc/init.rb

@@ -10,8 +10,12 @@ module IsoDoc
     end
 
     def i18n_init(lang, script, locale, i18nyaml = nil)
-      @i18n = I18n.new(lang, script, locale: locale,
-                                     i18nyaml: i18nyaml || @i18nyaml)
+      payload = i18nyaml || @i18nyaml
+      @i18n = if payload.is_a?(Hash)
+                I18n.new(lang, script, locale: locale, i18nhash: payload)
+              else
+                I18n.new(lang, script, locale: locale, i18nyaml: payload)
+              end
     end
 
     def l10n(expr, lang = @lang, script = @script, opt = {})
@@ -19,6 +23,13 @@ module IsoDoc
       @i18n.l10n(expr, lang, script, opt)
     end
 
+    def presentation_xml_converter
+      parts = self.class.name.split("::")
+      parts[-1] = "PresentationXMLConvert"
+      Object.const_get(parts.join("::"))
+        .new(language: @lang, script: @script)
+    end
+
     def docxml_var_init(docxml)
       doctype_init(docxml)
       toc_init(docxml)
@@ -27,8 +38,8 @@ module IsoDoc
     def doctype_init(docxml)
       @doctype = docxml.at(ns("//bibdata/ext/doctype"))&.text
       @subdoctype = docxml.at(ns("//bibdata/ext/subdoctype"))&.text
-      @docscheme = docxml.at(ns("//metanorma-extension/presentation-metadata/"\
-        "document-scheme"))&.text
+      @docscheme = docxml.at(ns("//metanorma-extension/presentation-metadata/" \
+                                "document-scheme"))&.text
     end
 
     def toc_init(docxml)
@@ -73,12 +84,12 @@ module IsoDoc
     def log_messages
       # rubocop:disable Naming/VariableNumber
       {
-        "STANDOC_36": { category: "Anchors",
-                        error: "ID %s has already been used at line %s",
-                        severity: 0 },
-        "ISODOC_1": { category: "Crossreferences",
-                      error: "Anchor %s pointed to by %s " \
-             "is not defined in the document", severity: 1 },
+        STANDOC_36: { category: "Anchors",
+                      error: "ID %s has already been used at line %s",
+                      severity: 0 },
+        ISODOC_1: { category: "Crossreferences",
+                    error: "Anchor %s pointed to by %s " \
+                           "is not defined in the document", severity: 1 },
       }
       # rubocop:enable Naming/VariableNumber
     end

data/lib/isodoc/presentation_function/concepts.rb

@@ -42,7 +42,7 @@ module IsoDoc
 
     def concept_dup_cleanup_orig(node)
       node.xpath(".//xmlns:semx[xmlns:fmt-xref | xmlns:fmt-eref | " \
-         "xmlns:fmt-origin | xmlns:fmt-link]").each(&:remove)
+                 "xmlns:fmt-origin | xmlns:fmt-link]").each(&:remove)
       node.xpath(ns(".//xref | .//eref | .//origin | .//link")).each do |x|
         x["original-id"] or next
         x["id"] = x["original-id"]

data/lib/isodoc/presentation_function/erefs.rb

@@ -6,12 +6,12 @@ module IsoDoc
     def citeas(xmldoc)
       xmldoc.xpath(ns("//fmt-eref | //fmt-origin | //fmt-link"))
         .each do |e|
-          sem_xml_descendant?(e) and next
-          e["bibitemid"] && e["citeas"] or next
-          a = @xrefs.anchor(e["bibitemid"], :xref, false) or next
-          e["citeas"] = citeas_cleanup(a)
-          # link generated in collection postprocessing from eref
-          e.name == "fmt-link" && e.text.empty? and e.children = e["citeas"]
+        sem_xml_descendant?(e) and next
+        (e["bibitemid"] && e["citeas"]) or next
+        a = @xrefs.anchor(e["bibitemid"], :xref, false) or next
+        e["citeas"] = citeas_cleanup(a)
+        # link generated in collection postprocessing from eref
+        e.name == "fmt-link" && e.text.empty? and e.children = e["citeas"]
       end
     end
 
@@ -42,24 +42,35 @@ module IsoDoc
     end
 
     def resolve_eref_connectives(locs)
+      locs = resolve_eref_connectives1(locs)
+      locs.size < 3 and return locs.map do |x|
+        x[:custom] || x[:conn] || x[:ref]
+      end
+      locs = resolve_eref_connectives_split(locs)
+      locs = locs.each_slice(2).with_object([]) do |a, m|
+        m << { custom: a[0][:custom], conn: a[0][:conn], label: a.dig(1, :ref) }
+      end
+      [", ", combine_conn(locs)]
+    end
+
+    def resolve_eref_connectives1(locs)
       locs = escape_l10n(locs)
       locs = resolve_comma_connectives(locs)
-      locs = resolve_to_connectives(locs)
-      locs.size < 3 and return locs.map { |x| x[:custom] || x[:conn] || x[:ref] }
-      locs = locs.each_with_object([]) do |x, m|
-        if m.empty? then m << x
+      resolve_to_connectives(locs)
+    end
+
+    def resolve_eref_connectives_split(locs)
+      locs.each_with_object([]) do |x, m|
+        if m.empty?
+          m << x
         elsif m[-1][:conn] && x[:conn]
           m[-1][:conn] += x[:conn]
           x[:custom] and m[-1][:custom] = x[:custom]
-        elsif  m[-1][:conn] && x[:conn]
+        elsif m[-1][:conn] && x[:conn]
           m[-1][:ref] += x[:ref]
-        else  m << x
+        else m << x
         end
       end
-      locs = locs.each_slice(2).with_object([]) do |a, m|
-        m << { custom: a[0][:custom], conn: a[0][:conn], label: a.dig(1, :ref) }
-      end
-      [", ", combine_conn(locs)]
     end
 
     XREF_CONNECTIVES = %w(from to or and).freeze
@@ -82,7 +93,8 @@ module IsoDoc
 
     def resolve_comma_connectives1(locs, locs1, add)
       if [", ", " ", ""].include?(locs.dig(1, :conn)) && locs.size > 2
-        add += [locs[0][:ref], locs[1][:custom] || locs[1][:conn], locs[2][:ref]].join
+        add += [locs[0][:ref], locs[1][:custom] || locs[1][:conn],
+                locs[2][:ref]].join
         locs.shift(3)
       else
         locs1 << add unless add.empty?
@@ -101,14 +113,12 @@ module IsoDoc
           c = locs[1][:custom] and x = conn_sub(x, c)
           locs1 << { ref: connectives_spans(x) }
           locs.shift(3)
+        elsif locs[0][:conn] == "from" && locs[0][:custom]
+          locs1 << { conn: locs[0][:custom] }
+          locs.shift # strip "from" and English
+        # TODO languages with obligatory "from"
         else
-          if locs[0][:conn] == "from" && locs[0][:custom] # strip "from" and English
-            # TODO languages with obligatory "from"
-            locs1 << { conn: locs[0][:custom] }
-            locs.shift
-          else
-            locs1 << locs.shift
-          end
+          locs1 << locs.shift
         end
       end
       locs1
@@ -118,14 +128,15 @@ module IsoDoc
       docxml.xpath(ns("//display-text")).each { |f| f.replace(f.children) }
       docxml.xpath(ns("//fmt-eref | //fmt-origin[not(.//termref)]"))
         .each do |e|
-          sem_xml_descendant?(e) and next
-          href = eref_target(e) or next
-          e.xpath(ns("./locality | ./localityStack")).each(&:remove)
-          if %w(short).include?(e["style"]) then eref2linkshort(e, href)
-          elsif href[:type] == :anchor || %w(full).include?(e["style"])
-            eref2xref(e)
-          else eref2link1(e, href)
-          end
+        sem_xml_descendant?(e) and next
+        href = eref_target(e) or next
+        e.xpath(ns("./locality | ./localityStack")).each(&:remove)
+        if %w(short).include?(e["style"])
+          eref2linkshort(e, href)
+        elsif href[:type] == :anchor || %w(full).include?(e["style"])
+          eref2xref(e)
+        else eref2link1(e, href)
+        end
       end
     end
 

data/lib/isodoc/presentation_function/math.rb

@@ -44,20 +44,24 @@ module IsoDoc
     def implicit_number_formatter(num, locale)
       num.ancestors("formula").empty? or return
       ## by default, no formatting in formulas
-      fmt = { significant: num_totaldigits(num.text) }.compact
+      fmt = { significant: num_totaldigits(num.text, 10) }.compact
       n = normalise_number(num.text)
       @numfmt.localized_number(n, locale:, format: fmt,
                                   precision: num_precision(num.text))
     end
 
     def numberformat_type(ret)
-      %i(precision significant digit_count group_digits fraction_group_digits)
+      %i(precision significant digit_count group_digits fraction_group_digits
+         base)
         .each do |i|
-          ret[i] &&= ret[i].to_i
-        end
-      %i(notation exponent_sign number_sign locale).each do |i|
+        ret[i] &&= ret[i].to_i
+      end
+      %i(notation exponent_sign number_sign locale hex_capital).each do |i|
         ret[i] &&= ret[i].to_sym
       end
+      %i(base_prefix base_suffix).each do |i|
+        ["", "nil"].include?(ret[i]) and ret[i] = nil
+      end
       ret
     end
 
@@ -76,14 +80,17 @@ module IsoDoc
         v.is_a?(String) ? HTMLEntities.new.decode(v) : v
       end.merge(fmt)
       symbols = large_notation_fmt(symbols, num.text)
-      [symbols[:precision] || num_precision(num.text), symbols,
-       explicit_number_formatter_signif(num, symbols)]
+      significant = explicit_number_formatter_signif(num, symbols)
+      precision = symbols[:precision] || num_precision(num.text) ||
+        num_precision_from_significant(num.text, significant,
+                                       (symbols[:base] || 10).to_i)
+      [precision, symbols, significant]
     end
 
     def explicit_number_formatter_signif(num, symbols)
       signif = symbols[:significant]
       (symbols.keys & %i(precision digit_count)).empty? and
-        signif ||= num_totaldigits(num.text)
+        signif ||= num_totaldigits(num.text, (symbols[:base] || 10).to_i)
       signif
     end
 
@@ -117,16 +124,56 @@ module IsoDoc
       precision
     end
 
-    def num_totaldigits(num)
+    def num_totaldigits(num, base = 10)
       totaldigits = nil
       /\.(?=\d+e)/.match?(num) and
         totaldigits = twitter_cldr_localiser_symbols[:significant] ||
-          # [^.]* and [^e]* exclude their respective delimiters,
-          # preventing polynomial backtracking.
-          num.sub(/\A0\./, ".").sub(/\A[^.]*\./, "").sub(/e[^e]*\z/, "").size
+          num_totaldigits_compute(num, base)
       totaldigits
     end
 
+    # In base 10, total significant digits = source mantissa length.
+    # In other bases, the converted value may have fewer digits. Per
+    # https://github.com/metanorma/isodoc/issues/788: if the source has
+    # no fractional part after E-notation expansion, return the
+    # length of the integer in the target base (e.g. 123 -> 7B = 2
+    # digits, not 3); if the source has a fractional part, preserve
+    # the source mantissa length so Plurimath pads the converted
+    # fraction (e.g. 123.25 -> 7B.400, 123.0 -> 7B.00).
+    def num_totaldigits_compute(num, base)
+      # [^.]* and [^e]* exclude their respective delimiters,
+      # preventing polynomial backtracking.
+      mantissa = num.sub(/\A0\./, ".").sub(/\A[^.]*\./, "")
+        .sub(/e[^e]*\z/, "")
+      return mantissa.size if base == 10
+
+      m = num.match(/\A-?0?\.(\d+)e(-?\d+)\z/)
+      if m && m[1].length <= m[2].to_i
+        BigDecimal(num).to_i.abs.to_s(base).length
+      else
+        mantissa.size
+      end
+    end
+
+    # When base != 10 and the source is in E-notation, derive the
+    # target-base fractional digit count from the total significant
+    # digit count we computed. Plurimath's default precision_from()
+    # returns the decimal fractional digit count, which
+    # Numbers::Fraction#change_base then uses as the target-base
+    # fractional digit count -- producing too few fractional digits
+    # in non-decimal bases. Workaround pending upstream fix; see
+    # https://github.com/metanorma/isodoc/issues/788.
+    def num_precision_from_significant(num, significant, base)
+      base == 10 and return nil
+      significant.nil? and return nil
+      /\.(?=\d+e)/.match?(num) or return nil
+      m = num.match(/\A-?0?\.(\d+)e(-?\d+)\z/) or return nil
+      m[1].length <= m[2].to_i and return 0
+      val = BigDecimal(num).abs
+      integer_len = val.to_i.zero? ? 0 : val.to_i.to_s(base).length
+      [significant - integer_len, 0].max
+    end
+
     def twitter_cldr_localiser_symbols
       {}
     end

data/lib/isodoc/version.rb

@@ -1,3 +1,3 @@
 module IsoDoc
-  VERSION = "3.5.0".freeze
+  VERSION = "3.5.1".freeze
 end

data/lib/isodoc/word_function/body.rb

@@ -13,7 +13,17 @@ module IsoDoc
         html.add_namespace("m", "http://schemas.microsoft.com/office/2004/12/omml")
       end
 
+      # Apply empty_tags even when postprocess is skipped (debug mode),
+      # so convert1's output is valid HTML5 (non-void self-closing tags
+      # expanded: <a id="_"/> -> <a id="_"></a>). Idempotent: running
+      # again in postprocess#toHTML is a no-op.
+      def convert1(docxml, filename, dir)
+        empty_tags(super)
+      end
+
       def define_head(head, filename, _dir)
+        head.meta "http-equiv": "Content-Type",
+                  content: "text/html; charset=UTF-8"
         head.style do |style|
           loc = File.join(File.dirname(__FILE__), "..", "base_style",
                           "metanorma_word.scss")

data/lib/isodoc/word_function/postprocess.rb

@@ -20,7 +20,7 @@ module IsoDoc
         result = from_xhtml(word_cleanup(result))
           .gsub("-DOUBLE_HYPHEN_ESCAPE-", "--")
         # Unescape & to & in href attributes after all Nokogiri processing
-        unescape_amp_in_hrefs(result)
+        unescape_amp_in_hrefs(empty_tags(result))
       end
 
       def toWord(result, filename, dir, header)

data/lib/isodoc/word_function/postprocess_cover.rb

@@ -41,7 +41,7 @@ module IsoDoc
         xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office"
         xmlns:w="urn:schemas-microsoft-com:office:word"
         xmlns:m="http://schemas.microsoft.com/office/2004/12/omml">
-        <head> <title></title> <meta charset="UTF-8" /> </head>
+        <head> <title></title> <meta charset="UTF-8"/> </head>
         <body> </body> </html>
       HERE
 

data/lib/isodoc/xref.rb

@@ -17,6 +17,7 @@ module IsoDoc
     include Function::Utils
 
     attr_reader :klass
+    attr_accessor :reqt_models
 
     # Note: if bibrender is no passed in, do not parse references
     def initialize(lang, script, klass, i18n, options = {})
@@ -27,9 +28,6 @@ module IsoDoc
       @options = options
       initialize_i18n(i18n)
       @klass.bibrender ||= options[:bibrender]
-      @reqt_models = @klass.requirements_processor
-        .new({ default: "default", lang:, script:,
-               labels: @i18n.get })
     end
 
     def initialize_empty

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: isodoc
 version: !ruby/object:Gem::Version
-  version: 3.5.0
+  version: 3.5.1
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-04-13 00:00:00.000000000 Z
+date: 2026-04-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: base64
@@ -192,6 +192,34 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: bigdecimal
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: canon
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '='
+      - !ruby/object:Gem::Version
+        version: 0.2.3
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '='
+      - !ruby/object:Gem::Version
+        version: 0.2.3
 - !ruby/object:Gem::Dependency
   name: debug
   requirement: !ruby/object:Gem::Requirement
@@ -332,34 +360,6 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '0.9'
-- !ruby/object:Gem::Dependency
-  name: canon
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - '='
-      - !ruby/object:Gem::Version
-        version: 0.1.3
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - '='
-      - !ruby/object:Gem::Version
-        version: 0.1.3
-- !ruby/object:Gem::Dependency
-  name: openssl
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
 description: |
   isodoc converts documents in the IsoDoc document model into
   Microsoft Word and HTML.