metanorma-standoc 1.10.5 → 1.11.0

data/lib/asciidoctor/standoc/cleanup_terms.rb

@@ -1,19 +1,9 @@
 require_relative "term_lookup_cleanup"
+require_relative "cleanup_terms_designations"
 
 module Asciidoctor
   module Standoc
     module Cleanup
-      def termdef_stem_cleanup(xmldoc)
-        xmldoc.xpath("//term/p/stem").each do |a|
-          if a.parent.elements.size == 1 # para contains just a stem expression
-            t = Nokogiri::XML::Element.new("admitted", xmldoc)
-            parent = a.parent
-            t.children = a.remove
-            parent.replace(t)
-          end
-        end
-      end
-
       def termdomain_cleanup(xmldoc)
         xmldoc.xpath("//p/domain").each do |a|
           prev = a.parent.previous
@@ -22,43 +12,53 @@ module Asciidoctor
       end
 
       def termdomain1_cleanup(xmldoc)
-        xmldoc.xpath("//domain").each do |d|
-          defn = d.at("../definition") and
-            defn.previous = d.remove
+        xmldoc.xpath("//term").each do |t|
+          d = t.xpath("./domain | ./subject | ./usageinfo").last or next
+          defn = d.at("../definition") and defn.previous = d.remove
         end
       end
 
       def termdefinition_cleanup(xmldoc)
-        xmldoc.xpath("//term").each do |d|
-          first_child = d.at("./p | ./figure | ./formula") || next
+        generate_termdefinitions(xmldoc)
+        split_termdefinitions(xmldoc)
+      end
+
+      def generate_termdefinitions(xmldoc)
+        xmldoc.xpath("//term[not(definition)]").each do |d|
+          first_child = d.at("./p | ./figure | ./formula | ./table") || next
           t = Nokogiri::XML::Element.new("definition", xmldoc)
           first_child.replace(t)
           t << first_child.remove
-          d.xpath("./p | ./figure | ./formula").each { |n| t << n.remove }
+          d.xpath("./p | ./figure | ./formula | ./table").each { |n| t << n.remove }
         end
       end
 
-      def termdef_unnest_cleanup(xmldoc)
-        # release termdef tags from surrounding paras
-        nodes = xmldoc.xpath("//p/admitted | //p/deprecates")
-        while !nodes.empty?
-          nodes[0].parent.replace(nodes[0].parent.children)
-          nodes = xmldoc.xpath("//p/admitted | //p/deprecates")
+      def split_termdefinitions(xmldoc)
+        xmldoc.xpath("//definition").each do |d|
+          n = d.children.first.add_previous_sibling("<nonverbalrepresentation/>").first
+          v = d.children.first.add_previous_sibling("<verbaldefinition/>").first
+          nonverb = false
+          d.elements.each do |e|
+            case e.name
+            when "nonverbalrepresentation", "verbaldefinition" then next
+            when "figure", "table", "formula"
+              n << e.remove
+              nonverb = true
+            when "termsource"
+              (nonverb ? n : v) << e.remove
+            else v << e.remove
+            end
+          end
         end
       end
 
-      def termdef_subclause_cleanup(xmldoc)
-        xmldoc.xpath("//terms[terms]").each { |t| t.name = "clause" }
-      end
-
       def termdocsource_cleanup(xmldoc)
         f = xmldoc.at("//preface | //sections")
-        xmldoc.xpath("//termdocsource").each do |s|
-          f.previous = s.remove
-        end
+        xmldoc.xpath("//termdocsource").each { |s| f.previous = s.remove }
       end
 
       def term_children_cleanup(xmldoc)
+        xmldoc.xpath("//terms[terms]").each { |t| t.name = "clause" }
         xmldoc.xpath("//term").each do |t|
           %w(termnote termexample termsource).each do |w|
             t.xpath("./#{w}").each { |n| t << n.remove }
@@ -75,69 +75,29 @@ module Asciidoctor
       end
 
       def termnote_example_cleanup(xmldoc)
-        xmldoc.xpath("//termnote[not(ancestor::term)]").each do |x|
-          x.name = "note"
+        %w(note example).each do |w|
+        xmldoc.xpath("//term#{w}[not(ancestor::term)]").each do |x|
+          x.name = w
         end
-        xmldoc.xpath("//termexample[not(ancestor::term)]").each do |x|
-          x.name = "example"
         end
       end
 
       def termdef_cleanup(xmldoc)
+        termdef_unnest_cleanup(xmldoc)
         Asciidoctor::Standoc::TermLookupCleanup.new(xmldoc, @log).call
+        term_nonverbal_designations(xmldoc)
+        term_dl_to_metadata(xmldoc)
+        term_termsource_to_designation(xmldoc)
+        term_designation_reorder(xmldoc)
         termdef_from_termbase(xmldoc)
-        termdef_unnest_cleanup(xmldoc)
         termdef_stem_cleanup(xmldoc)
         termdomain_cleanup(xmldoc)
         termdefinition_cleanup(xmldoc)
         termdomain1_cleanup(xmldoc)
         termnote_example_cleanup(xmldoc)
-        termdef_subclause_cleanup(xmldoc)
         term_children_cleanup(xmldoc)
         termdocsource_cleanup(xmldoc)
       end
-
-      # Indices sort after letter but before any following
-      # letter (x, x_m, x_1, xa); we use colon to force that sort order.
-      # Numbers sort *after* letters; we use thorn to force that sort order.
-      def symbol_key(sym)
-        key = sym.dup
-        key.traverse do |n|
-          next unless n.name == "math"
-
-          n.replace(grkletters(MathML2AsciiMath.m2a(n.to_xml)))
-        end
-        ret = Nokogiri::XML(key.to_xml)
-        HTMLEntities.new.decode(ret.text.downcase)
-          .gsub(/[\[\]{}<>()]/, "").gsub(/\s/m, "")
-          .gsub(/[[:punct:]]|[_^]/, ":\\0").gsub(/`/, "")
-          .gsub(/[0-9]+/, "þ\\0")
-      end
-
-      def grkletters(x)
-        x.gsub(/\b(alpha|beta|gamma|delta|epsilon|zeta|eta|theta|iota|kappa|lambda|mu|nu|xi|omicron|pi|rho|sigma|tau|upsilon|phi|chi|psi|omega)\b/i, "&\\1;")
-      end
-
-      def extract_symbols_list(dlist)
-        dl_out = []
-        dlist.xpath("./dt | ./dd").each do |dtd|
-          if dtd.name == "dt"
-            dl_out << { dt: dtd.remove, key: symbol_key(dtd) }
-          else
-            dl_out.last[:dd] = dtd.remove
-          end
-        end
-        dl_out
-      end
-
-      def symbols_cleanup(docxml)
-        docxml.xpath("//definitions/dl").each do |dl|
-          dl_out = extract_symbols_list(dl)
-          dl_out.sort! { |a, b| a[:key] <=> b[:key] || a[:dt] <=> b[:dt] }
-          dl.children = dl_out.map { |d| d[:dt].to_s + d[:dd].to_s }.join("\n")
-        end
-        docxml
-      end
     end
   end
 end

data/lib/asciidoctor/standoc/cleanup_terms_designations.rb

@@ -0,0 +1,162 @@
+module Asciidoctor
+  module Standoc
+    module Cleanup
+      def termdef_stem_cleanup(xmldoc)
+        xmldoc.xpath("//term/p/stem").each do |a|
+          if a.parent.elements.size == 1 # para contains just a stem expression
+            parent = a.parent
+            parent.replace("<admitted>#{term_expr(a.to_xml)}</admitted>")
+          end
+        end
+        xmldoc.xpath("//term//expression/name[stem]").each do |n|
+          n.parent.name = "letter-symbol"
+        end
+      end
+
+      # release termdef tags from surrounding paras
+      def termdef_unnest_cleanup(xmldoc)
+        desgn = "//p/admitted | //p/deprecates | //p/preferred | //p//related"
+        nodes = xmldoc.xpath(desgn)
+        while !nodes.empty?
+          nodes[0].parent.replace(nodes[0].parent.children)
+          nodes = xmldoc.xpath(desgn)
+        end
+      end
+
+      def term_dl_to_metadata(xmldoc)
+        xmldoc.xpath("//term[dl[@metadata = 'true']]").each do |t|
+          t.xpath("./dl[@metadata = 'true']").each do |dl|
+            prev = dl_to_designation(dl) or next
+            term_dl_to_designation_metadata(prev, dl)
+            term_dl_to_term_metadata(prev, dl)
+            term_dl_to_expression_metadata(prev, dl)
+            dl.remove
+          end
+        end
+      end
+
+      def term_dl_to_term_metadata(prev, dlist)
+        return unless prev.name == "preferred" &&
+          prev.at("./preceding-sibling::preferred").nil?
+
+        ins = term_element_insert_point(prev)
+        %w(domain subject usageinfo).each do |a|
+          ins = dl_to_elems(ins, prev.parent, dlist, a)
+        end
+      end
+
+      def term_dl_to_designation_metadata(prev, dlist)
+        %w(absent geographicArea).each { |a| dl_to_attrs(related2pref(prev), dlist, a) }
+      end
+
+      def term_element_insert_point(prev)
+        ins = prev
+        while %w(preferred admitted deprecates related domain dl)
+            .include? ins&.next_element&.name
+          ins = ins.next_element
+        end
+        ins
+      end
+
+      def term_dl_to_expression_metadata(prev, dlist)
+        %w(language script type isInternational).each do |a|
+          dl_to_attrs(prev, dlist, a)
+        end
+        %w(abbreviationType pronunciation).reverse.each do |a|
+          dl_to_elems(prev.at("./expression/name"), prev, dlist, a)
+        end
+        g = dlist.at("./dt[text()='grammar']/following::dd//dl") and
+          term_dl_to_expression_grammar(prev, g)
+        term_to_letter_symbol(prev, dlist)
+      end
+
+      def term_dl_to_expression_grammar(prev, dlist)
+        prev.at(".//expression") or return
+        prev.at(".//expression") << "<grammar><sentinel/></grammar>"
+        %w(gender isPreposition isParticiple isAdjective isAdverb isNoun
+           grammarValue).reverse.each do |a|
+          dl_to_elems(prev.at(".//expression/grammar/*"), prev.elements.last,
+                      dlist, a)
+        end
+        term_dl_to_designation_gender(prev)
+      end
+
+      def term_dl_to_designation_gender(prev)
+        gender = prev.at(".//expression/grammar/gender")
+        /,/.match?(gender&.text) and
+          gender.replace(gender.text.split(/,\s*/)
+            .map { |x| "<gender>#{x}</gender>" }.join)
+        prev.at(".//expression/grammar/sentinel").remove
+      end
+
+      def term_to_letter_symbol(prev, dlist)
+        ls = dlist.at("./dt[text()='letter-symbol']/following::dd/p")
+        return unless ls&.text == "true"
+
+        prev.at(".//expression").name = "letter-symbol"
+      end
+
+      def dl_to_designation(dlist)
+        prev = dlist.previous_element
+        unless %w(preferred admitted deprecates related).include? prev&.name
+          @log.add("AsciiDoc Input", dlist, "Metadata definition list does "\
+                                            "not follow a term designation")
+          return nil
+        end
+        prev
+      end
+
+      def term_nonverbal_designations(xmldoc)
+        xmldoc.xpath("//term/preferred | //term/admitted | //term/deprecates")
+          .each do |d|
+          d.text.strip.empty? or next
+          n = d.next_element
+          if %w(formula figure).include?(n&.name)
+            term_nonverbal_designations1(d, n)
+          else d.at("./expression/name") or
+            d.children = term_expr("")
+          end
+        end
+      end
+
+      def term_nonverbal_designations1(desgn, elem)
+        desgn = related2pref(desgn)
+        if elem.name == "figure"
+          elem.at("./name").remove
+          desgn.children =
+            "<graphical-symbol>#{elem.remove.to_xml}</graphical-symbol>"
+        else
+          desgn.children = term_expr(elem.at("./stem").to_xml)
+          elem.remove
+        end
+      end
+
+      def term_termsource_to_designation(xmldoc)
+        xmldoc.xpath("//term/termsource").each do |t|
+          p = t.previous_element
+          while %w(domain subject usageinfo).include? p&.name
+            p = p.previous_element
+          end
+          %w(preferred admitted deprecates related).include?(p&.name) or
+            next
+          related2pref(p) << t.remove
+        end
+      end
+
+      def term_designation_reorder(xmldoc)
+        xmldoc.xpath("//term").each do |t|
+          %w(preferred admitted deprecates related)
+            .each_with_object([]) do |tag, m|
+            t.xpath("./#{tag}").each { |x| m << x.remove }
+          end.reverse.each do |x|
+            t.children.first.previous = x
+          end
+        end
+      end
+
+      def related2pref(elem)
+        elem.name == "related" ? elem = elem.at("./preferred") : elem
+      end
+    end
+  end
+end

data/lib/asciidoctor/standoc/cleanup_text.rb

@@ -9,7 +9,8 @@ module Asciidoctor
                   </passthrough>}mx) { HTMLEntities.new.decode($1) }
       end
 
-            IGNORE_DUMBQUOTES = "//pre | //pre//* | //tt | //tt//* | "\
+      IGNORE_DUMBQUOTES =
+        "//pre | //pre//* | //tt | //tt//* | "\
         "//sourcecode | //sourcecode//* | //bibdata//* | //stem | "\
         "//stem//* | //figure[@class = 'pseudocode'] | "\
         "//figure[@class = 'pseudocode']//*".freeze
@@ -32,6 +33,7 @@ module Asciidoctor
                      "[starts-with(., '\"') or starts-with(., \"'\")]]")
           .each do |x|
           next if !x.ancestors("pre, tt, sourcecode, stem, figure").empty?
+
           uninterrupt_quotes_around_xml1(x)
         end
       end
@@ -40,7 +42,8 @@ module Asciidoctor
         prev = elem.at(".//preceding::text()[1]") or return
         /\S$/.match?(prev.text) or return
         foll = elem.at(".//following::text()[1]")
-        m = /^(["'][[:punct:]]*)(\s|$)/.match(HTMLEntities.new.decode(foll&.text)) or return
+        m = /^(["'][[:punct:]]*)(\s|$)/
+          .match(HTMLEntities.new.decode(foll&.text)) or return
         foll.content = foll.text.sub(/^(["'][[:punct:]]*)/, "")
         prev.content = "#{prev.text}#{m[1]}"
       end

data/lib/asciidoctor/standoc/cleanup_xref.rb

@@ -0,0 +1,107 @@
+module Asciidoctor
+  module Standoc
+    module Cleanup
+      # extending localities to cover ISO referencing
+      LOCALITY_REGEX_STR = <<~REGEXP.freeze
+        ^((?<locality>section|clause|part|paragraph|chapter|page|
+                      table|annex|figure|example|note|formula|list|time|anchor|
+                      locality:[^ \\t\\n\\r:,;=]+)(\\s+|=)
+               (?<ref>[^"][^ \\t\\n,:-]*|"[^"]+")
+                 (-(?<to>[^"][^ \\t\\n,:-]*|"[^"]"))?|
+          (?<locality2>whole|locality:[^ \\t\\n\\r:,;=]+))(?<punct>[,:;]?)\\s*
+         (?<text>.*)$
+      REGEXP
+      LOCALITY_RE = Regexp.new(LOCALITY_REGEX_STR.gsub(/\s/, ""),
+                               Regexp::IGNORECASE | Regexp::MULTILINE)
+
+      def tq(text)
+        text.sub(/^"/, "").sub(/"$/, "")
+      end
+
+      def extract_localities(elem)
+        f = elem&.children&.first or return
+        f.text? or return
+        head = f.remove.text
+        tail = elem&.children&.remove
+        extract_localities1(elem, head)
+        tail and elem << tail
+      end
+
+      def extract_localities1(elem, text)
+        b = elem.add_child("<localityStack/>").first if LOCALITY_RE.match text
+        while (m = LOCALITY_RE.match text)
+          ref = m[:ref] ? "<referenceFrom>#{tq m[:ref]}</referenceFrom>" : ""
+          refto = m[:to] ? "<referenceTo>#{tq m[:to]}</referenceTo>" : ""
+          b.add_child("<locality type='#{locality_label(m)}'>#{ref}#{refto}"\
+                      "</locality>")
+          text = m[:text]
+          b = elem.add_child("<localityStack/>").first if m[:punct] == ";"
+        end
+        elem.add_child(text) if text
+      end
+
+      def locality_label(match)
+        loc = match[:locality] || match[:locality2]
+        /^locality:/.match?(loc) ? loc : loc&.downcase
+      end
+
+      def xref_to_eref(elem)
+        elem["bibitemid"] = elem["target"]
+        unless elem["citeas"] = @anchors&.dig(elem["target"], :xref)
+          @internal_eref_namespaces.include?(elem["type"]) or
+            @log.add("Crossreferences", elem,
+                     "#{elem['target']} does not have a corresponding "\
+                     "anchor ID in the bibliography!")
+        end
+        elem.delete("target")
+        extract_localities(elem) unless elem.children.empty?
+      end
+
+      def xref_cleanup(xmldoc)
+        xmldoc.xpath("//xref").each do |x|
+          /:/.match(x["target"]) and xref_to_internal_eref(x)
+          next unless x.name == "xref"
+
+          if refid? x["target"]
+            x.name = "eref"
+            xref_to_eref(x)
+          else x.delete("type")
+          end
+        end
+      end
+
+      def xref_to_internal_eref(elem)
+        a = elem["target"].split(":", 3)
+        unless a.size < 2 || a[0].empty? || a[1].empty?
+          elem["target"] = "#{a[0]}_#{a[1]}"
+          a.size > 2 and
+            elem.children = %{anchor="#{a[2..-1].join}",#{elem&.children&.text}}
+          elem["type"] = a[0]
+          @internal_eref_namespaces << a[0]
+          elem.name = "eref"
+          xref_to_eref(elem)
+        end
+      end
+
+      def quotesource_cleanup(xmldoc)
+        xmldoc.xpath("//quote/source | //terms/source").each do |x|
+          xref_to_eref(x)
+        end
+      end
+
+      def origin_cleanup(xmldoc)
+        xmldoc.xpath("//origin/concept[termref]").each do |x|
+          t = x.at("./termref")
+          x.replace(t)
+        end
+        xmldoc.xpath("//origin").each do |x|
+          x["citeas"] = @anchors&.dig(x["bibitemid"], :xref) or
+            @log.add("Crossreferences", x,
+                     "#{x['bibitemid']} does not have a corresponding anchor "\
+                     "ID in the bibliography!")
+          extract_localities(x) unless x.children.empty?
+        end
+      end
+    end
+  end
+end

data/lib/asciidoctor/standoc/converter.rb

@@ -28,8 +28,10 @@ module Asciidoctor
         preprocessor Metanorma::Plugin::Lutaml::LutamlPreprocessor
         preprocessor Metanorma::Plugin::Lutaml::LutamlUmlAttributesTablePreprocessor
         preprocessor Metanorma::Plugin::Lutaml::LutamlUmlDatamodelDescriptionPreprocessor
+        inline_macro Asciidoctor::Standoc::PreferredTermInlineMacro
         inline_macro Asciidoctor::Standoc::AltTermInlineMacro
         inline_macro Asciidoctor::Standoc::DeprecatedTermInlineMacro
+        inline_macro Asciidoctor::Standoc::RelatedTermInlineMacro
         inline_macro Asciidoctor::Standoc::DomainTermInlineMacro
         inline_macro Asciidoctor::Standoc::InheritInlineMacro
         inline_macro Asciidoctor::Standoc::HTML5RubyMacro
@@ -49,7 +51,9 @@ module Asciidoctor
         inline_macro Asciidoctor::Standoc::FormSelectMacro
         inline_macro Asciidoctor::Standoc::FormOptionMacro
         inline_macro Asciidoctor::Standoc::ToCInlineMacro
+        inline_macro Asciidoctor::Standoc::PassInlineMacro
         inline_macro Metanorma::Plugin::Lutaml::LutamlFigureInlineMacro
+        inline_macro Metanorma::Plugin::Lutaml::LutamlTableInlineMacro
         block_macro Metanorma::Plugin::Lutaml::LutamlDiagramBlockMacro
         block Asciidoctor::Standoc::ToDoAdmonitionBlock
         treeprocessor Asciidoctor::Standoc::ToDoInlineAdmonitionBlock
@@ -97,6 +101,17 @@ module Asciidoctor
         File.join(@libdir, "../../isodoc/html", file)
       end
 
+      def content(node)
+        node.content
+      end
+
+      def skip(node, name = nil)
+        name = name || node.node_name
+        w = "converter missing for #{name} node in Metanorma backend"
+        @log.add("AsciiDoc Input", node, w)
+        nil
+      end
+
       alias_method :embedded, :content
       alias_method :verse, :quote
       alias_method :audio, :skip

data/lib/asciidoctor/standoc/inline.rb

@@ -170,8 +170,10 @@ module Asciidoctor
           else
             case node.role
               # the following three are legacy, they are now handled by macros
-            when "alt" then xml.admitted { |a| a << node.text }
-            when "deprecated" then xml.deprecates { |a| a << node.text }
+            when "alt"
+              term_designation(xml, node, "admitted", node.text)
+            when "deprecated"
+              term_designation(xml, node, "deprecates", node.text)
             when "domain" then xml.domain { |a| a << node.text }
 
             when "strike" then xml.strike { |s| s << node.text }
@@ -209,7 +211,7 @@ module Asciidoctor
       def inline_image(node)
         noko do |xml|
           xml.image **image_attributes(node)
-        end.join("")
+        end.join
       end
 
       def inline_indexterm(node)
@@ -218,8 +220,8 @@ module Asciidoctor
           terms = (node.attr("terms") || [node.text]).map { |x| xml_encode(x) }
           xml.index do |i|
             i.primary { |x| x << terms[0] }
-            a = terms.dig(1) and i.secondary { |x| x << a }
-            a = terms.dig(2) and i.tertiary { |x| x << a }
+            a = terms[1] and i.secondary { |x| x << a }
+            a = terms[2] and i.tertiary { |x| x << a }
           end
         end.join
       end