metanorma-standoc 2.2.5 → 2.2.7

data/lib/metanorma/standoc/cleanup_biblio.rb

@@ -1,113 +1,6 @@
-require "set"
-require "relaton_bib"
-
 module Metanorma
   module Standoc
     module Cleanup
-      def ref_dl_cleanup(xmldoc)
-        xmldoc.xpath("//clause[@bibitem = 'true']").each do |c|
-          bib = dl_bib_extract(c) or next
-          validate_ref_dl(bib, c)
-          bibitemxml = RelatonBib::BibliographicItem.from_hash(bib).to_xml or next
-          bibitem = Nokogiri::XML(bibitemxml)
-          bibitem.root["id"] = c["id"] if c["id"] && !/^_/.match(c["id"])
-          c.replace(bibitem.root)
-        end
-      end
-
-      # do not accept implicit id
-      def validate_ref_dl(bib, clause)
-        id = bib["id"]
-        id ||= clause["id"] unless /^_/.match?(clause["id"])
-        unless id
-          @log.add("Anchors", clause,
-                   "The following reference is missing an anchor:\n"\
-                   "#{clause.to_xml}")
-          return
-        end
-        @refids << id
-        validate_ref_dl1(bib, id, clause)
-      end
-
-      def validate_ref_dl1(bib, id, clause)
-        bib["title"] or
-          @log.add("Bibliography", clause, "Reference #{id} is missing a title")
-        bib["docid"] or
-          @log.add("Bibliography", clause,
-                   "Reference #{id} is missing a document identifier (docid)")
-      end
-
-      def extract_from_p(tag, bib, key)
-        return unless bib[tag]
-
-        "<#{key}>#{bib[tag].at('p').children}</#{key}>"
-      end
-
-      # if the content is a single paragraph, replace it with its children
-      # single links replaced with uri
-      def p_unwrap(para)
-        elems = para.elements
-        if elems.size == 1 && elems[0].name == "p"
-          link_unwrap(elems[0]).children.to_xml.strip
-        else
-          para.to_xml.strip
-        end
-      end
-
-      def link_unwrap(para)
-        elems = para.elements
-        if elems.size == 1 && elems[0].name == "link"
-          para.at("./link").replace(elems[0]["target"].strip)
-        end
-        para
-      end
-
-      def dd_bib_extract(dtd)
-        return nil if dtd.children.empty?
-
-        dtd.at("./dl") and return dl_bib_extract(dtd)
-        elems = dtd.remove.elements
-        return p_unwrap(dtd) unless elems.size == 1 &&
-          %w(ol ul).include?(elems[0].name)
-
-        elems[0].xpath("./li").each_with_object([]) do |li, ret|
-          ret << p_unwrap(li)
-        end
-      end
-
-      def add_to_hash(bib, key, val)
-        Metanorma::Utils::set_nested_value(bib, key.split("."), val)
-      end
-
-      # definition list, with at most one level of unordered lists
-      def dl_bib_extract(clause, nested = false)
-        dl = clause.at("./dl") or return
-        key = ""
-        bib = dl.xpath("./dt | ./dd").each_with_object({}) do |dtd, m|
-          (dtd.name == "dt" and key = dtd.text.sub(/:+$/, "")) and next
-          add_to_hash(m, key, dd_bib_extract(dtd))
-        end
-        clause.xpath("./clause").each do |c1|
-          key = c1&.at("./title")&.text&.downcase&.strip
-          next unless %w(contributor relation series).include? key
-
-          add_to_hash(bib, key, dl_bib_extract(c1, true))
-        end
-        dl_bib_extract_title(bib, clause, nested)
-      end
-
-      def dl_bib_extract_title(bib, clause, nested)
-        (!nested && clause.at("./title")) or return bib
-        title = clause.at("./title").remove.children.to_xml
-        bib["title"] = [bib["title"]] if bib["title"].is_a?(Hash) ||
-          bib["title"].is_a?(String)
-        bib["title"] ||= []
-        bib["title"] << title if !title.empty?
-        bib
-      end
-
-      # ---
-
       def formattedref_spans(xmldoc)
         xmldoc.xpath("//bibitem[formattedref//span]").each do |b|
           spans = spans_preprocess(extract_content(b))
@@ -156,7 +49,8 @@ module Metanorma
       def span_preprocess1(span, ret)
         case span[:key]
         when "uri", "docid"
-          ret[span[:key].to_sym] << { type: span[:type], val: span[:val] }
+          val = link_unwrap(Nokogiri::XML.fragment(span[:val])).to_xml
+          ret[span[:key].to_sym] << { type: span[:type], val: val }
         when "date"
           ret[span[:key].to_sym] << { type: span[:type] || "published",
                                       val: span[:val] }
@@ -263,10 +157,11 @@ module Metanorma
 
       def span_to_extent(span, key)
         values = span.split(/[-–]/)
-        ret = "<extent type='#{key}'><referenceFrom>#{values[0]}</referenceFrom>"
+        ret = "<extent type='#{key}'><locality>" \
+              "<referenceFrom>#{values[0]}</referenceFrom>"
         values[1] and
           ret += "<referenceTo>#{values[1]}</referenceTo>"
-        "#{ret}</extent>"
+        "#{ret}</locality></extent>"
       end
 
       def span_to_docid(span, key)
@@ -287,10 +182,10 @@ module Metanorma
 
       def span_to_person(span)
         pre = (span[:"formatted-initials"] and
-                     "<formatted-initials>"\
+                     "<formatted-initials>" \
                      "#{span[:"formatted-initials"]}</formatted-initials>") ||
           "<forename>#{span[:givenname]}</forename>"
-        "<person><name>#{pre}<surname>#{span[:surname]}</surname></name>"\
+        "<person><name>#{pre}<surname>#{span[:surname]}</surname></name>" \
           "</person>"
       end
     end

data/lib/metanorma/standoc/cleanup_block.rb

@@ -7,7 +7,8 @@ module Metanorma
       def para_cleanup(xmldoc)
         ["//p[not(ancestor::bibdata)]", "//ol[not(ancestor::bibdata)]",
          "//ul[not(ancestor::bibdata)]", "//quote[not(ancestor::bibdata)]",
-         "//note[not(ancestor::bibitem or "\
+         "//dl[not(ancestor::bibdata)]",
+         "//note[not(ancestor::bibitem or " \
          "ancestor::table or ancestor::bibdata)]"].each do |w|
           inject_id(xmldoc, w)
         end
@@ -106,7 +107,7 @@ module Metanorma
           prev = n.previous_element || next
           n.parent = prev if ELEMS_ALLOW_NOTES.include? prev.name
         end
-        xmldoc.xpath("//note[@keep-separate] | "\
+        xmldoc.xpath("//note[@keep-separate] | " \
                      "//termnote[@keep-separate]").each do |n|
           n.delete("keep-separate")
         end
@@ -182,10 +183,10 @@ module Metanorma
       end
 
       def block_index_cleanup(xmldoc)
-        xmldoc.xpath("//quote | //td | //th | //formula | //li | //dt | "\
-                     "//dd | //example | //note | //figure | //sourcecode | "\
-                     "//admonition | //termnote | //termexample | //form  | "\
-                     "//requirement | //recommendation | //permission | "\
+        xmldoc.xpath("//quote | //td | //th | //formula | //li | //dt | " \
+                     "//dd | //example | //note | //figure | //sourcecode | " \
+                     "//admonition | //termnote | //termexample | //form  | " \
+                     "//requirement | //recommendation | //permission | " \
                      "//imagemap | //svgmap").each do |b|
           b.xpath("./p[indexterm]").each do |p|
             indexterm_para?(p) or next

data/lib/metanorma/standoc/cleanup_boilerplate.rb

@@ -53,11 +53,11 @@ module Metanorma
         end
       end
 
-      TERM_CLAUSE = "//sections/terms | "\
+      TERM_CLAUSE = "//sections/terms | " \
                     "//sections/clause[descendant::terms]".freeze
 
       NORM_REF =
-        "//bibliography/references[@normative = 'true'][not(@hidden)] | "\
+        "//bibliography/references[@normative = 'true'][not(@hidden)] | " \
         "//bibliography/clause[.//references[@normative = 'true']]".freeze
 
       def boilerplate_isodoc(xmldoc)
@@ -79,7 +79,7 @@ module Metanorma
 
       def unwrap_boilerplate_clauses(xmldoc, xpath)
         xmldoc.xpath(xpath).each do |f|
-          f.xpath(".//clause[@type = 'boilerplate'] | "\
+          f.xpath(".//clause[@type = 'boilerplate'] | " \
                   ".//note[@type = 'boilerplate']").each do |c|
             c&.at("./title")&.remove
             c.replace(c.children)
@@ -175,9 +175,9 @@ module Metanorma
       end
 
       def indirect_eref_to_xref(eref, ident)
-        loc = eref&.at("./localityStack[locality[@type = 'anchor']]")
+        loc = eref.at("./localityStack[locality[@type = 'anchor']]")
           &.remove&.text ||
-          eref&.at("./locality[@type = 'anchor']")&.remove&.text || ident
+          eref.at("./locality[@type = 'anchor']")&.remove&.text || ident
         eref.name = "xref"
         eref.delete("bibitemid")
         eref.delete("citeas")
@@ -213,7 +213,7 @@ module Metanorma
       def bibdata_embed_hdr_cleanup(xmldoc)
         return if @embed_hdr.nil? || @embed_hdr.empty?
 
-        xmldoc.at("//bibdata") << "<relation type='derivedFrom'>"\
+        xmldoc.at("//bibdata") << "<relation type='derivedFrom'>" \
                                   "#{hdr2bibitem(@embed_hdr.first)}</relation>"
       end
 

data/lib/metanorma/standoc/cleanup_inline.rb

@@ -35,7 +35,7 @@ module Metanorma
 
       def li_bookmark_cleanup(xmldoc)
         xmldoc.xpath("//li[descendant::bookmark]").each do |x|
-          if x.at("./*[1][local-name() = 'p']/"\
+          if x.at("./*[1][local-name() = 'p']/" \
                   "*[1][local-name() = 'bookmark']") &&
               empty_text_before_first_element(x.elements[0])
             bookmark_to_id(x, x.elements[0].elements[0])
@@ -45,7 +45,7 @@ module Metanorma
 
       def dt_bookmark_cleanup(xmldoc)
         xmldoc.xpath("//dt[descendant::bookmark]").each do |x|
-          if x.at("./*[1][local-name() = 'p']/"\
+          if x.at("./*[1][local-name() = 'p']/" \
                   "*[1][local-name() = 'bookmark']") &&
               empty_text_before_first_element(x.elements[0])
             bookmark_to_id(x, x.elements[0].elements[0])

data/lib/metanorma/standoc/cleanup_maths.rb

@@ -8,16 +8,24 @@ module Metanorma
           "<amathstem>#{@c.decode($1)}</amathstem>"
         end
         text = Html2Doc.new({})
-          .asciimath_to_mathml(text, ["<amathstem>", "</amathstem>"])
-        x =  Nokogiri::XML(text)
+          .asciimath_to_mathml(text, ["<amathstem>", "</amathstem>"],
+                               retain_asciimath: true)
+        asciimath2mathml_wrap(text)
+      end
+
+      def asciimath2mathml_wrap(text)
+        x = Nokogiri::XML(text)
         x.xpath("//*[local-name() = 'math'][not(parent::stem)]").each do |y|
           y.wrap("<stem type='MathML'></stem>")
         end
-        x.to_xml
+        x.xpath("//stem").each do |y|
+          y.next_element&.name == "asciimath" and y << y.next_element
+        end
+        to_xml(x)
       end
 
       def xml_unescape_mathml(xml)
-        return if xml.children.any? { |y| y.element? }
+        return if xml.children.any?(&:element?)
 
         math = xml.text.gsub(/&lt;/, "<").gsub(/&gt;/, ">")
           .gsub(/&quot;/, '"').gsub(/&apos;/, "'").gsub(/&amp;/, "&")

data/lib/metanorma/standoc/cleanup_section.rb

@@ -8,7 +8,7 @@ module Metanorma
   module Standoc
     module Cleanup
       def make_preface(xml, sect)
-        if xml.at("//foreword | //introduction | //acknowledgements | "\
+        if xml.at("//foreword | //introduction | //acknowledgements | " \
                   "//*[@preface]")
           preface = sect.add_previous_sibling("<preface/>").first
           f = xml.at("//foreword") and preface.add_child f.remove
@@ -33,21 +33,24 @@ module Metanorma
           abstract = xml.at("//abstract[not(ancestor::bibitem)]").remove
           preface.prepend_child abstract.remove
           bibabstract = bibabstract_location(xml)
-          dupabstract = abstract.dup
-          dupabstract.traverse { |n| n.remove_attribute("id") }
-          dupabstract.remove_attribute("language")
-          dupabstract.remove_attribute("script")
-          dupabstract&.at("./title")&.remove
-          bibabstract.next = dupabstract
+          bibabstract.next = clean_abstract(abstract.dup)
         end
       end
 
+      def clean_abstract(dupabstract)
+        dupabstract.traverse { |n| n.remove_attribute("id") }
+        dupabstract.remove_attribute("language")
+        dupabstract.remove_attribute("script")
+        dupabstract.at("./title")&.remove
+        dupabstract
+      end
+
       def bibabstract_location(xml)
         xml.at("//bibdata/script") || xml.at("//bibdata/language") ||
           xml.at("//bibdata/contributor[not(following-sibling::contributor)]") ||
           xml.at("//bibdata/date[not(following-sibling::date)]") ||
           xml.at("//docnumber") ||
-          xml.at("//bibdata/docidentifier"\
+          xml.at("//bibdata/docidentifier" \
                  "[not(following-sibling::docidentifier)]") ||
           xml.at("//bibdata/uri[not(following-sibling::uri)]") ||
           xml.at("//bibdata/title[not(following-sibling::title)]")
@@ -111,6 +114,7 @@ module Metanorma
       end
 
       def sections_cleanup(xml)
+        misccontainer_cleanup(xml)
         sections_order_cleanup(xml)
         sections_level_cleanup(xml)
         sections_names_cleanup(xml)
@@ -118,6 +122,12 @@ module Metanorma
         change_clauses(xml)
       end
 
+      def misccontainer_cleanup(xml)
+        m = xml.at("//misc-container-clause") or return
+        ins = add_misc_container(xml)
+        ins << m.remove.children
+      end
+
       def single_clause_annex(xml)
         xml.xpath("//annex").each do |a|
           single_clause_annex1(a)
@@ -131,7 +141,7 @@ module Metanorma
       end
 
       def obligations_cleanup_info(xml)
-        xml.xpath("//foreword | //introduction | //acknowledgements | "\
+        xml.xpath("//foreword | //introduction | //acknowledgements | " \
                   "//references | //preface//clause").each do |r|
           r["obligation"] = "informative"
         end

data/lib/metanorma/standoc/cleanup_symbols.rb

@@ -5,18 +5,25 @@ module Metanorma
       # 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|
-          n.name == "math" and
-            n.replace(grkletters(MathML2AsciiMath.m2a(n.to_xml)))
-        end
-        ret = Nokogiri::XML(key.to_xml)
-        @c.decode(ret.text.downcase)
+        @c.decode(asciimath_key(sym).text.downcase)
           .gsub(/[\[\]{}<>()]/, "").gsub(/\s/m, "")
           .gsub(/[[:punct:]]|[_^]/, ":\\0").gsub(/`/, "")
           .gsub(/[0-9]+/, "þ\\0")
       end
 
+      def asciimath_key(sym)
+        key = sym.dup
+        key.traverse do |n|
+          if n.name == "math"
+            n.children = @c.encode(
+              @c.decode(grkletters(MathML2AsciiMath.m2a(n.to_xml))), :basic
+            )
+          end
+        end
+        key.xpath(".//asciimath").each(&:remove)
+        Nokogiri::XML(key.to_xml)
+      end
+
       def grkletters(text)
         text.gsub(/\b(alpha|beta|gamma|delta|epsilon|zeta|eta|theta|iota|kappa|
                       lambda|mu|nu|xi|omicron|pi|rho|sigma|tau|upsilon|phi|chi|

data/lib/metanorma/standoc/cleanup_text.rb

@@ -35,7 +35,8 @@ module Metanorma
       end
 
       IGNORE_QUOTES_ELEMENTS =
-        %w(pre tt sourcecode stem figure bibdata passthrough identifier).freeze
+        %w(pre tt sourcecode stem asciimath figure bibdata passthrough
+           identifier).freeze
 
       def uninterrupt_quotes_around_xml_skip(elem)
         !(/\A['"]/.match?(elem.text) &&
@@ -43,7 +44,7 @@ module Metanorma
           .intersection(IGNORE_QUOTES_ELEMENTS).empty? &&
           ((elem.previous.text.strip.empty? &&
             !empty_tag_with_text_content?(elem.previous)) ||
-           elem.previous.name == "index"))
+           ignoretext?(elem.previous)))
       end
 
       def uninterrupt_quotes_around_xml1(elem)
@@ -56,13 +57,20 @@ module Metanorma
         prev.content = "#{prev.text}#{m[1]}"
       end
 
+      IGNORE_TEXT_ELEMENTS =
+        %w(index fn).freeze
+
+      def ignoretext?(elem)
+        IGNORE_TEXT_ELEMENTS.include? elem.name
+      end
+
       def block?(elem)
         %w(title name variant-title clause figure annex example introduction
            foreword acknowledgements note li th td dt dd p quote label
            abstract preferred admitted related deprecates field-of-application
            usage-info expression pronunciation grammar-value domain
            definition termnote termexample modification description
-           newcontent floating-title).include? elem.name
+           newcontent floating-title tab).include? elem.name
       end
 
       def empty_tag_with_text_content?(elem)
@@ -79,7 +87,7 @@ module Metanorma
           ancestors = x.path.split(%r{/})[1..-2]
           ancestors.intersection(IGNORE_QUOTES_ELEMENTS).empty? and
             dumb2smart_quotes1(x, prev)
-          prev = x.text unless ancestors.include?("index")
+          prev = x.text if ancestors.intersection(IGNORE_TEXT_ELEMENTS).empty?
         end
       end
 
@@ -93,9 +101,14 @@ module Metanorma
 
       def dumbquote_cleanup(xmldoc)
         xmldoc.traverse do |n|
-          next unless n.text?
+          next unless n.text? && /\u2019/.match?(n.text)
 
-          n.replace(n.text.gsub(/(?<=\p{Alnum})\u2019(?=\p{Alpha})/, "'")) # .
+          # n.replace(n.text.gsub(/(?<=\p{Alnum})\u2019(?=\p{Alpha})/, "'")) # .
+          n.replace(@c.encode(
+                      @c.decode(n.text)
+            .gsub(/(?<=\p{Alnum})\u2019(?=\p{Alpha})/, "'"),
+                      :basic, :hexadecimal
+                    ))
         end
       end
     end

data/lib/metanorma/standoc/cleanup_toc.rb

@@ -80,8 +80,7 @@ module Metanorma
       def toc_metadata(xmldoc)
         return unless @htmltoclevels || @doctoclevels || @toclevels
 
-        ins = xmldoc.at("//misc-container") ||
-          xmldoc.at("//bibdata").after("<misc-container/>").next_element
+        ins = add_misc_container(xmldoc)
         toc_metadata1(ins)
       end
 

data/lib/metanorma/standoc/cleanup_xref.rb

@@ -20,10 +20,10 @@ module Metanorma
       end
 
       def extract_localities(elem)
-        f = elem&.children&.first or return
+        f = elem.children&.first or return
         f.text? or return
         head = f.remove.text
-        tail = elem&.children&.remove
+        tail = elem.children&.remove
         extract_localities1(elem, head)
         tail and elem << tail
       end
@@ -45,7 +45,7 @@ module Metanorma
         ref =
           match[:ref] ? "<referenceFrom>#{tq match[:ref]}</referenceFrom>" : ""
         refto = match[:to] ? "<referenceTo>#{tq match[:to]}</referenceTo>" : ""
-        stack.add_child("<locality type='#{locality_label(match)}'>#{ref}"\
+        stack.add_child("<locality type='#{locality_label(match)}'>#{ref}" \
                         "#{refto}</locality>")
       end
 
@@ -67,7 +67,8 @@ module Metanorma
         /^locality:/.match?(loc) ? loc : loc&.downcase
       end
 
-      def xref_to_eref(elem)
+      def xref_to_eref(elem, name)
+        elem.name = name
         elem["bibitemid"] = elem["target"]
         if ref = @anchors&.dig(elem["target"], :xref)
           elem["citeas"] = @c.decode(ref)
@@ -82,14 +83,49 @@ module Metanorma
       def xref_to_eref1(elem)
         @internal_eref_namespaces.include?(elem["type"]) or
           @log.add("Crossreferences", elem,
-                   "#{elem['target']} does not have a corresponding "\
+                   "#{elem['target']} does not have a corresponding " \
                    "anchor ID in the bibliography!")
       end
 
       def xref_cleanup(xmldoc)
+        anchor_alias(xmldoc)
         xref_compound_cleanup(xmldoc)
         xref_cleanup1(xmldoc)
         xref_compound_wrapup(xmldoc)
+        eref_stack(xmldoc)
+      end
+
+      def eref_stack(xmldoc)
+        xmldoc.xpath("//eref[eref]").each do |e|
+          e.name = "erefstack"
+          e.delete("bibitemid")
+          e.delete("citeas")
+          e.xpath("./eref").each do |e1|
+            e1["type"] = e["type"]
+          end
+          e.delete("type")
+        end
+      end
+
+      def anchor_alias(xmldoc)
+        t = xmldoc.at("//misc-container/table[@id = " \
+                      "'_misccontainer_anchor_aliases']") or return
+        key = ""
+        t.xpath("./tbody/tr").each do |tr|
+          tr.xpath("./td | ./th").each_with_index do |td, i|
+            if i.zero? then key = td.text
+            else anchor_alias1(key, td)
+            end
+          end
+        end
+      end
+
+      def anchor_alias1(key, elem)
+        id = elem.text.strip
+        id.empty? and elem.at("./link") and
+          id = elem.at("./link/@target")&.text
+        (key && !id.empty?) or return
+        @anchor_alias[id] = key
       end
 
       def xref_compound_cleanup(xmldoc)
@@ -128,12 +164,12 @@ module Metanorma
 
       def xref_cleanup1(xmldoc)
         xmldoc.xpath("//xref").each do |x|
-          /:/.match?(x["target"]) and xref_to_internal_eref(x)
+          %r{:(?!//)}.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)
+            xref_to_eref(x, "eref")
+          elsif @anchor_alias[x["target"]] then xref_alias(x)
           else x.delete("type")
           end
         end
@@ -144,17 +180,22 @@ module Metanorma
         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.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)
+          xref_to_eref(elem, "eref")
         end
       end
 
+      def xref_alias(elem)
+        elem["style"] == "id" && elem.text.strip.empty? and
+          elem << elem["target"]
+        elem["target"] = @anchor_alias[elem["target"]]
+      end
+
       def quotesource_cleanup(xmldoc)
         xmldoc.xpath("//quote/source | //terms/source").each do |x|
-          xref_to_eref(x)
+          xref_to_eref(x, "source")
         end
       end
 
@@ -165,9 +206,9 @@ module Metanorma
         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 "\
+                     "#{x['bibitemid']} does not have a corresponding anchor " \
                      "ID in the bibliography!")
-          extract_localities(x) unless x.children.empty?
+          x.children.empty? or extract_localities(x)
         end
       end
     end

data/lib/metanorma/standoc/inline.rb

@@ -40,28 +40,44 @@ module Metanorma
       end
 
       def inline_anchor_xref_attrs(node)
-        m = inline_anchor_xref_match(node)
+        text = concatenate_attributes_to_xref_text(node)
+        m = inline_anchor_xref_match(text)
         t = node.target.gsub(/^#/, "").gsub(%r{(\.xml|\.adoc)(#.*$)}, "\\2")
-        m.nil? and return { target: t, type: "inline", text: node.text }
-        { target: t, type: m[:fn].nil? ? "inline" : "footnote",
-          case: m[:case]&.sub(/%$/, ""),
-          droploc: m[:drop].nil? && m[:drop2].nil? ? nil : true,
-          text: inline_anchor_xref_text(m, node),
-          hidden: m[:hidden] }
+        m.nil? and return { target: t, type: "inline", text: text }
+        inline_anchor_xref_attrs1(m, t, text)
       end
 
-      def inline_anchor_xref_match(node)
+      def concatenate_attributes_to_xref_text(node)
+        node.attributes.each_with_object([]) do |(k, v), m|
+          next if %w(path fragment refid).include?(k)
+
+          m << "#{k}=#{v}%"
+        end.map { |x| x.sub(/%+/, "%") }.join + (node.text || "")
+      end
+
+      def inline_anchor_xref_attrs1(match, target, text)
+        { target: target,
+          type: match[:fn].nil? ? "inline" : "footnote",
+          case: match[:case]&.sub(/%$/, ""),
+          style: match[:style]&.sub(/^style=/, "")&.sub(/%$/, "") || @xrefstyle,
+          droploc: match[:drop].nil? && match[:drop2].nil? ? nil : true,
+          text: inline_anchor_xref_text(match, text),
+          hidden: match[:hidden] }
+      end
+
+      def inline_anchor_xref_match(text)
         /^(?:hidden%(?<hidden>[^,]+),?)?
+          (?<style>style=[^%]+%)?
           (?<drop>droploc%)?(?<case>capital%|lowercase%)?(?<drop2>droploc%)?
-          (?<fn>fn:?\s*)?(?<text>.*)$/x.match node.text
+          (?<fn>fn:?\s*)?(?<text>.*)$/x.match text
       end
 
-      def inline_anchor_xref_text(match, node)
-        if %i[case fn drop drop2 hidden].any? do |x|
+      def inline_anchor_xref_text(match, text)
+        if %i[case fn drop drop2 hidden style].any? do |x|
              !match[x].nil?
            end
           match[:text]
-        else node.text
+        else text
         end
       end
 
@@ -158,6 +174,7 @@ module Metanorma
             .elements[0]
           math.delete("alttext")
           s.parent.children = math
+          s << "<latexmath>#{text}</latexmath>"
         end
       end
 

data/lib/metanorma/standoc/isodoc-compile.rng

@@ -0,0 +1,7 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<grammar ns="https://www.metanorma.org/ns/standoc" xmlns="http://relaxng.org/ns/structure/1.0">
+  <!-- default namespace isostandard = "https://www.metanorma.com/ns/iso" -->
+  <include href="biblio-standoc.rng"/>
+  <include href="isodoc.rng"/>
+</grammar>
+<!-- end overrides -->