metanorma-standoc 2.0.2 → 2.0.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: d9031c28646f9cd9a481ad67911f2f0f7e5bd058368f80888a788753e5f63fa3
-  data.tar.gz: 25b16459f351f266f19b2f00d70b5bf7f6f5c95325a0d61d43fbe5f3fa518802
+  metadata.gz: cc4b55c7be36492191bd86bc19646a520c76e94bd2e2af9f96cf0d80d25fb05e
+  data.tar.gz: 533fc5380552ab239528a559ee41a47c246203e17cb33df64cb5cdef5df21f34
 SHA512:
-  metadata.gz: 885e62405f3198482ffd7e152b25bef8ab09fc227a8107984297e8fa15b64552b2e745f4ce5e1d60dd37df6dc5cf5c505fb9acfeedd1bdc4ef77954e01218893
-  data.tar.gz: 69cac251a0696b46dd25f37fb23828780e1945e57464e829c91ab5d88ac481311d130c4ff0e670ee89d56fd998eff4ea48e3c5a0c4b24c514021431fb785d852
+  metadata.gz: adc2d0b10dd30e5f59060b20574701dc6049b25a9d25eff06557ac51ca66645f62b553517690380009460f01fe135d26e83ab653b81c7746209e8f55221c3703
+  data.tar.gz: 8bc78cf1ab42e80d93c8a22f0028e44e7ba735914fa43c2dcdd5eb53cb57b3a3ee52ea1677cae5033dbc738f8b444bc0dc2e5021ff08a443d991535c84c373b2

data/lib/metanorma/standoc/base.rb

@@ -71,7 +71,7 @@ module Metanorma
 
       def document(node)
         init(node)
-        ret = makexml(node).to_xml(encoding: "US-ASCII", indent: 2)
+        ret = makexml(node).to_xml(encoding: "UTF-8", indent: 2)
         outputs(node, ret) unless node.attr("nodoc") || !node.attr("docfile")
         clean_exit
         ret

data/lib/metanorma/standoc/basicdoc.rng

@@ -173,9 +173,11 @@
           <data type="dateTime"/>
         </attribute>
       </optional>
-      <attribute name="from">
-        <data type="IDREF"/>
-      </attribute>
+      <optional>
+        <attribute name="from">
+          <data type="IDREF"/>
+        </attribute>
+      </optional>
       <optional>
         <attribute name="to">
           <data type="IDREF"/>

data/lib/metanorma/standoc/biblio.rng

@@ -209,9 +209,6 @@
       <zeroOrMore>
         <ref name="contact"/>
       </zeroOrMore>
-      <zeroOrMore>
-        <ref name="uri"/>
-      </zeroOrMore>
     </element>
   </define>
   <define name="fullname">
@@ -828,6 +825,11 @@
       <optional>
         <attribute name="scope"/>
       </optional>
+      <optional>
+        <attribute name="primary">
+          <data type="boolean"/>
+        </attribute>
+      </optional>
       <text/>
     </element>
   </define>

data/lib/metanorma/standoc/cleanup_image.rb

@@ -2,6 +2,7 @@ module Metanorma
   module Standoc
     module Cleanup
       def svgmap_cleanup(xmldoc)
+        svg_uniqueids(xmldoc)
         svgmap_moveattrs(xmldoc)
         svgmap_populate(xmldoc)
         Metanorma::Utils::svgmap_rewrite(xmldoc, @localdir)
@@ -59,12 +60,125 @@ module Metanorma
       end
 
       def img_cleanup(xmldoc)
-        return xmldoc unless @datauriimage
+        if @datauriimage
+          xmldoc.xpath("//image").each do |i|
+            # do not datauri encode SVG, we need to deduplicate its IDs
+            unless read_in_if_svg(i, @localdir)
+              i["src"] = Metanorma::Utils::datauri(i["src"], @localdir)
+            end
+          end
+        end
+        svg_uniqueids(xmldoc)
+        xmldoc
+      end
+
+      def read_in_if_svg(img, localdir)
+        return false unless img["src"]
+
+        path = Metanorma::Utils::svgmap_rewrite0_path(img["src"], localdir)
+        File.file?(path) or return false
+        types = MIME::Types.type_for(path) or return false
+        types.first == "image/svg+xml" or return false
+        svg = File.read(path, encoding: "utf-8") or return false
+        img.replace(Nokogiri::XML(svg).root.to_xml)
+        true
+      end
+
+      IRI_TAG_PROPERTIES_MAP = {
+        clipPath: ["clip-path"],
+        "color-profile": nil,
+        cursor: nil,
+        filter: nil,
+        linearGradient: ["fill", "stroke"],
+        marker: ["marker", "marker-end", "marker-mid", "marker-start"],
+        mask: nil,
+        pattern: ["fill", "stroke"],
+        radialGradient: ["fill", "stroke"],
+      }.freeze
+
+      SVG_NS = "http://www.w3.org/2000/svg".freeze
+
+      def svg_uniqueids(xmldoc)
+        # only keep non-unique identifiers
+        ids = xmldoc.xpath("//m:svg//*/@id | //svg/@id", "m" => SVG_NS)
+          .map(&:text).group_by(&:itself).transform_values(&:count)
+          .delete_if { |_, v| v < 2 }
+        xmldoc.xpath("//m:svg", "m" => SVG_NS).each_with_index do |s, i|
+          ids = svg_uniqueids1(s, i, ids)
+        end
+      end
+
+      def svg_iri_properties(id_elems)
+        iri_tag_names = id_elems.each_with_object([]) do |e, m|
+          IRI_TAG_PROPERTIES_MAP.key?(e.name.to_sym) and m = m << e.name
+        end.uniq
+        iri_properties = iri_tag_names.each_with_object([]) do |t, m|
+          (IRI_TAG_PROPERTIES_MAP[t.to_sym] || [t]).each { |t1| m = m << t1 }
+        end.uniq
+        return [] if iri_properties.empty?
+
+        iri_properties << "style"
+      end
 
-        xmldoc.xpath("//image").each do |i|
-          i["src"] = Metanorma::Utils::datauri(i["src"], @localdir)
+      def svg_uniqueids1(svg, idx, ids)
+        id_elems = svg.xpath(".//*[@id] | ./@id/..")
+        iri_properties = svg_iri_properties(id_elems)
+        svg_uniqueids2(svg, iri_properties, idx, ids)
+        new_ids = id_elems.map { |x| x["id"] }
+          .map { |x| x + (ids[x] ? "_inject_#{idx}" : "") }
+        ids.merge(new_ids.each.map { |value| [value, true] }.to_h)
+      end
+
+      def svg_uniqueids2(svg, iri_properties, idx, ids)
+        svg.traverse do |e|
+          next unless e.element?
+
+          if e.name == "style"
+            svg_styleupdate(e, idx, ids)
+          elsif !e.attributes.empty?
+            svg_attrupdate(e, iri_properties, idx, ids)
+          end
+          svg_linkupdate(e, idx, ids)
+          svg_idupdate(e, idx, ids)
         end
       end
+
+      def svg_update_url(text, idx, ids)
+        text.gsub(/url\("?#([a-zA-Z][\w:.-]*)"?\)/) do |x|
+          if ids[$1] then "url(##{$1}_inject_#{idx})"
+          else x
+          end
+        end
+      end
+
+      def svg_styleupdate(elem, idx, ids)
+        elem.children = svg_update_url(elem.text, idx, ids)
+      end
+
+      def svg_attrupdate(elem, iri_properties, idx, ids)
+        iri_properties.each do |p|
+          next unless elem[p]
+
+          elem[p] = svg_update_url(elem[p], idx, ids)
+        end
+      end
+
+      def svg_linkupdate(elem, idx, ids)
+        %w(xlink:href href).each do |ref|
+          iri = elem[ref]&.strip
+          next unless /^#/.match?(iri)
+          next unless ids[iri.sub(/^#/, "")]
+
+          elem[ref] += "_inject_#{idx}"
+        end
+      end
+
+      def svg_idupdate(elem, idx, ids)
+        return unless elem["id"]
+        return unless ids[elem["id"]]
+
+        elem["id"] += "_inject_#{idx}"
+      end
     end
   end
 end

data/lib/metanorma/standoc/cleanup_ref.rb

@@ -117,7 +117,7 @@ module Metanorma
           # isopub = ref.at(ISO_PUBLISHER_XPATH)
           docid = ref.at("./docidentifier[@type = 'metanorma']") ||
             ref.at("./docidentifier[not(@type = 'DOI')]") or next
-          reference = format_ref(docid.text, docid["type"])
+          reference = format_ref(docid.children.to_xml, docid["type"])
           @anchors[ref["id"]] = { xref: reference }
         end
       end

data/lib/metanorma/standoc/cleanup_section.rb

@@ -193,7 +193,7 @@ module Metanorma
             next unless t.next_element.nil?
             next if %w(sections annex preface).include? t.parent.name
 
-            t.parent.parent << t
+            t.parent.next = t
             found = true
           end
           break unless found

data/lib/metanorma/standoc/cleanup_xref.rb

@@ -47,16 +47,23 @@ module Metanorma
 
       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!")
+        if ref = @anchors&.dig(elem["target"], :xref)
+          elem["citeas"] = HTMLEntities.new.encode(ref, :hexadecimal)
+        else
+          elem["citeas"] = ""
+          xref_to_eref1(elem)
         end
         elem.delete("target")
         extract_localities(elem) unless elem.children.empty?
       end
 
+      def xref_to_eref1(elem)
+        @internal_eref_namespaces.include?(elem["type"]) or
+          @log.add("Crossreferences", elem,
+                   "#{elem['target']} does not have a corresponding "\
+                   "anchor ID in the bibliography!")
+      end
+
       def xref_cleanup(xmldoc)
         xmldoc.xpath("//xref").each do |x|
           /:/.match(x["target"]) and xref_to_internal_eref(x)

data/lib/metanorma/standoc/isodoc.rng

@@ -1098,6 +1098,16 @@
     </define>
   </include>
   <!-- end overrides -->
+  <define name="image" combine="choice">
+    <element name="svg">
+      <oneOrMore>
+        <choice>
+          <text/>
+          <ref name="AnyElement"/>
+        </choice>
+      </oneOrMore>
+    </element>
+  </define>
   <define name="MultilingualRenderingType">
     <choice>
       <value>common</value>

data/lib/metanorma/standoc/ref.rb

@@ -131,6 +131,7 @@ module Metanorma
       # TODO: alternative where only title is available
       def refitemcode(item, node)
         m = NON_ISO_REF.match(item) and return refitem1code(item, m).compact
+        m = NON_ISO_REF1.match(item) and return refitem1code(item, m).compact
         @log.add("AsciiDoc Input", node, "#{MALFORMED_REF}: #{item}")
         {}
       end
@@ -180,7 +181,11 @@ module Metanorma
         (<fn[^>]*>\s*<p>(?<fn>[^\]]+)</p>\s*</fn>,?\s?)?(?<text>.*)$}xm.freeze
 
       NON_ISO_REF = %r{^<ref\sid="(?<anchor>[^"]+)">
-      \[(?<usrlbl>\([^)]+\))?(?<code>[^\]]+?)\]</ref>,?\s*(?<text>.*)$}xm
+      \[(?<usrlbl>\([^)]+\))?(?<code>.+?)\]</ref>,?\s*(?<text>.*)$}xm
+        .freeze
+
+      NON_ISO_REF1 = %r{^<ref\sid="(?<anchor>[^"]+)">
+      (?<usrlbl>\([^)]+\))?(?<code>.+?)</ref>,?\s*(?<text>.*)$}xm
         .freeze
 
       def reference1_matches(item)

data/lib/metanorma/standoc/ref_utility.rb

@@ -75,7 +75,8 @@ module Metanorma
 
       def analyse_ref_repo_path(ret)
         return ret unless m =
-                            /^(?<type>repo|path):\((?<key>[^,]+),?(?<id>.*)\)$/.match(ret[:id])
+                            /^(?<type>repo|path):\((?<key>[^,]+),?(?<id>.*)\)$/
+                              .match(ret[:id])
 
         id = m[:id].empty? ? m[:key].sub(%r{^[^/]+/}, "") : m[:id]
         ret.merge(id: id, type: m[:type], key: m[:key], nofetch: true)

data/lib/metanorma/standoc/term_lookup_cleanup.rb

@@ -58,7 +58,7 @@ module Metanorma
 
       def set_termxref_tags_target
         xmldoc.xpath("//termxref").each do |node|
-          target = normalize_ref_id(node.text)
+          target = normalize_ref_id(node)
           if termlookup[:term][target].nil? && termlookup[:symbol][target].nil?
             remove_missing_ref(node, target)
             next
@@ -78,7 +78,7 @@ module Metanorma
 
       def remove_missing_ref_term(node, target)
         log.add("AsciiDoc Input", node,
-                %(Error: Term reference in `term[#{target}]` missing: \
+                %(Error: Term reference to `#{target}` missing: \
                 "#{target}" is not defined in document))
         node.name = "strong"
         node&.at("../xrefrender")&.remove
@@ -141,7 +141,7 @@ module Metanorma
       end
 
       def normalize_id_and_memorize_init(node, res_table, text_selector, prefix)
-        term_text = normalize_ref_id(node.at(text_selector).text)
+        term_text = normalize_ref_id(node.at(text_selector))
         unless AUTOMATIC_GENERATED_ID_REGEXP.match(node["id"]).nil? &&
             !node["id"].nil?
           id = unique_text_id(term_text, prefix)
@@ -155,12 +155,14 @@ module Metanorma
         node.xpath(text_selector).each_with_index do |p, i|
           next unless i.positive?
 
-          res_table[normalize_ref_id(p.text)] = node["id"]
+          res_table[normalize_ref_id(p)] = node["id"]
         end
       end
 
-      def normalize_ref_id(text)
-        Metanorma::Utils::to_ncname(text.downcase.gsub(/[[:space:]]/, "-"))
+      def normalize_ref_id(term)
+        t = term.dup
+        t.xpath(".//index").map(&:remove)
+        Metanorma::Utils::to_ncname(t.text.downcase.gsub(/[[:space:]]/, "-"))
       end
 
       def unique_text_id(text, prefix)

data/lib/metanorma/standoc/terms.rb

@@ -92,7 +92,7 @@ module Metanorma
         end
       end
 
-      def term_source_attrs(_node, seen_xref)
+      def termsource_origin_attrs(_node, seen_xref)
         { case: seen_xref.children[0]["case"],
           droploc: seen_xref.children[0]["droploc"],
           bibitemid: seen_xref.children[0]["target"],
@@ -103,7 +103,7 @@ module Metanorma
         if seen_xref.children[0].name == "concept"
           xml_t.origin { |o| o << seen_xref.children[0].to_xml }
         else
-          attrs = term_source_attrs(node, seen_xref)
+          attrs = termsource_origin_attrs(node, seen_xref)
           attrs.delete(:text)
           xml_t.origin **attr_code(attrs) do |o|
             o << seen_xref.children[0].children.to_xml
@@ -135,13 +135,16 @@ module Metanorma
         matched
       end
 
+      def termsource_attrs(node, matched)
+        status = node.attr("status") ||
+          (matched[:text] ? "modified" : "identical")
+        { status: status, type: node.attr("type") || "authoritative" }
+      end
+
       def termsource(node)
-        matched = extract_termsource_refs(node.content, node) || return
+        matched = extract_termsource_refs(node.content, node) or return
         noko do |xml|
-          status = node.attr("status") ||
-            (matched[:text] ? "modified" : "identical")
-          attrs = { status: status, type: node.attr("type") || "authoritative" }
-          xml.termsource **attrs do |xml_t|
+          xml.termsource **termsource_attrs(node, matched) do |xml_t|
             seen_xref = Nokogiri::XML.fragment(matched[:xref])
             add_term_source(node, xml_t, seen_xref, matched)
           end

data/lib/metanorma/standoc/validate.rb

@@ -120,20 +120,25 @@ module Metanorma
         end
       end
 
+      SVG_NS = "http://www.w3.org/2000/svg".freeze
+
       # RelaxNG cannot cope well with wildcard attributes. So we strip
       # any attributes from FormattedString instances (which can contain
       # xs:any markup, and are signalled with @format) before validation.
       def formattedstr_strip(doc)
         doc.xpath("//*[@format] | //stem | //bibdata//description | "\
                   "//formattedref | //bibdata//note | //bibdata/abstract | "\
-                  "//bibitem/abstract | //bibitem/note | //misc-container")
-          .each do |n|
+                  "//bibitem/abstract | //bibitem/note | //misc-container",
+                  "m" => SVG_NS).each do |n|
           n.elements.each do |e|
             e.traverse do |e1|
               e1.element? and e1.each { |k, _v| e1.delete(k) }
             end
           end
         end
+        doc.xpath("//m:svg", "m" => SVG_NS).each do |n|
+          n.replace("<svg/>")
+        end
         doc
       end
 

data/lib/metanorma/standoc/version.rb

@@ -19,6 +19,6 @@ module Metanorma
   end
 
   module Standoc
-    VERSION = "2.0.2".freeze
+    VERSION = "2.0.3".freeze
   end
 end

data/spec/metanorma/cleanup_blocks_spec.rb

@@ -1039,4 +1039,140 @@ RSpec.describe Metanorma::Standoc do
     annotation_id = output.at("//xmlns:annotation/@id").text
     expect(callout_id).to eq(annotation_id)
   end
+  it "deduplicates identifiers in inline SVGs" do
+    input = <<~INPUT
+      #{BLANK_HDR}
+        <sections>
+        <svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="0 0 256 256">
+         <defs>
+           <linearGradient id="gradient1">
+             <stop class="stop1" offset="0%" xlink:href="#gradient1"/>
+             <stop class="stop2" offset="100%"/>
+             <style>url(#gradient1)</style>
+           </linearGradient>
+         </defs>
+         <circle fill="url(#gradient1)" cx="128" cy="128" r="100" />
+       </svg>
+               <svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="0 0 256 256">
+         <defs>
+           <linearGradient id="gradient2">
+             <stop class="stop1" offset="0%" xlink:href="#gradient2"/>
+             <stop class="stop2" offset="100%"/>
+             <style>url(#gradient2)</style>
+           </linearGradient>
+         </defs>
+         <circle fill="url(#gradient2)" cx="128" cy="128" r="100" />
+       </svg>
+             <svg version="1.1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" viewBox="0 0 256 256">
+         <defs>
+           <linearGradient id="gradient1">
+             <stop class="stop1" offset="0%" xlink:href="#gradient1"/>
+             <stop class="stop2" offset="100%"/>
+             <style>url(#gradient1)</style>
+           </linearGradient>
+         </defs>
+         <circle fill="url(#gradient1)" cx="128" cy="128" r="100" />
+       </svg>
+             </sections>
+      </standard-document>
+    INPUT
+    output = <<~OUTPUT
+      #{BLANK_HDR}
+                 <sections>
+                   <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1" viewBox="0 0 256 256">
+         <defs>
+           <linearGradient id="gradient1_inject_0">
+             <stop class="stop1" offset="0%" xlink:href="#gradient1_inject_0"/>
+             <stop class="stop2" offset="100%"/>
+             <style>url(#gradient1_inject_0)</style>
+           </linearGradient>
+         </defs>
+         <circle fill="url(#gradient1_inject_0)" cx="128" cy="128" r="100"/>
+       </svg>
+               <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1" viewBox="0 0 256 256">
+         <defs>
+           <linearGradient id="gradient2">
+             <stop class="stop1" offset="0%" xlink:href="#gradient2"/>
+             <stop class="stop2" offset="100%"/>
+             <style>url(#gradient2)</style>
+           </linearGradient>
+         </defs>
+         <circle fill="url(#gradient2)" cx="128" cy="128" r="100"/>
+       </svg>
+             <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1" viewBox="0 0 256 256">
+         <defs>
+           <linearGradient id="gradient1_inject_2">
+             <stop class="stop1" offset="0%" xlink:href="#gradient1_inject_2"/>
+             <stop class="stop2" offset="100%"/>
+             <style>url(#gradient1_inject_2)</style>
+           </linearGradient>
+         </defs>
+         <circle fill="url(#gradient1_inject_2)" cx="128" cy="128" r="100"/>
+       </svg>
+             </sections>
+      </standard-document>
+    OUTPUT
+    expect(Metanorma::Standoc::Converter.new(nil, *OPTIONS)
+      .cleanup(Nokogiri::XML(input)).to_xml)
+      .to be_equivalent_to xmlpp(output)
+  end
+
+  it "deduplicates identifiers in embedded SVGs" do
+    input = <<~INPUT
+      #{ASCIIDOC_BLANK_HDR.sub(/:data-uri-image: false/, ':data-uri-image: true')}
+
+      image::spec/fixtures/action_schemaexpg1.svg[]
+
+      image::spec/examples/rice_images/rice_image1.png[]
+
+      image::spec/fixtures/action_schemaexpg1.svg[]
+    INPUT
+
+    output = <<~OUTPUT
+       #{BLANK_HDR}
+                <sections>
+          <figure id='_'>
+            <svg xmlns='http://www.w3.org/2000/svg' xmlns:xlink='http://www.w3.org/1999/xlink' version='1.1' id='Layer_1' x='0px' y='0px' viewBox='0 0 595.28 841.89' style='enable-background:new 0 0 595.28 841.89;' xml:space='preserve'>
+              <style/>
+              <image/>
+              <a xlink:href='mn://action_schema'>
+                <rect x='123.28' y='273.93' class='st0' width='88.05' height='41.84'/>
+              </a>
+              <a xlink:href='mn://basic_attribute_schema'>
+                <rect x='324.69' y='450.52' class='st0' width='132.62' height='40.75'/>
+              </a>
+              <a xlink:href='mn://support_resource_schema'>
+                <rect x='324.69' y='528.36' class='st0' width='148.16' height='40.75'/>
+              </a>
+            </svg>
+          </figure>
+          <figure id='_'>
+              <image/>
+          </figure>
+          <figure id='_'>
+            <svg xmlns='http://www.w3.org/2000/svg' xmlns:xlink='http://www.w3.org/1999/xlink' version='1.1' id='Layer_1_inject_1' x='0px' y='0px' viewBox='0 0 595.28 841.89' style='enable-background:new 0 0 595.28 841.89;' xml:space='preserve'>
+              <style/>
+              <image/>
+              <a xlink:href='mn://action_schema'>
+                <rect x='123.28' y='273.93' class='st0' width='88.05' height='41.84'/>
+              </a>
+              <a xlink:href='mn://basic_attribute_schema'>
+                <rect x='324.69' y='450.52' class='st0' width='132.62' height='40.75'/>
+              </a>
+              <a xlink:href='mn://support_resource_schema'>
+                <rect x='324.69' y='528.36' class='st0' width='148.16' height='40.75'/>
+              </a>
+            </svg>
+          </figure>
+        </sections>
+      </standard-document>
+    OUTPUT
+    xml = Nokogiri::XML(Asciidoctor.convert(input, *OPTIONS))
+    xml.xpath("//*[local-name() = 'image']").each do |x|
+      x.replace("<image/>")
+    end
+    expect(xmlpp(strip_guid(xml.to_xml)
+      .gsub(%r{<style.*?</style>}m, "<style/>")))
+      .to be_equivalent_to xmlpp(output)
+  end
 end

data/spec/metanorma/cleanup_spec.rb

@@ -585,7 +585,7 @@ RSpec.describe Metanorma::Standoc do
           <title type='main' format='text/plain' language='en' script='Latn'>International Electrotechnical Vocabulary (IEV) — Part 102: Mathematics — General concepts and linear algebra</title>
           <uri type="src">https://webstore.iec.ch/publication/160</uri>
           <uri type="obp">/preview/info_iec60050-102%7Bed1.0%7Db.pdf</uri>
-          <docidentifier type="IEC">IEC 60050-102:2007</docidentifier>
+          <docidentifier type="IEC" primary="true">IEC 60050-102:2007</docidentifier>
           <docidentifier type='URN'>urn:iec:std:iec:60050-102:2007:::en</docidentifier>
           <date type="published">
             <on>2007-08-27</on>
@@ -624,7 +624,7 @@ RSpec.describe Metanorma::Standoc do
           <title type="main" format="text/plain" language="en" script="Latn">International Electrotechnical Vocabulary (IEV) — Part 103: Mathematics — Functions</title>
           <uri type="src">https://webstore.iec.ch/publication/161</uri>
           <uri type="obp">/preview/info_iec60050-103%7Bed1.0%7Db.pdf</uri>
-          <docidentifier type="IEC">IEC 60050-103:2009</docidentifier>
+          <docidentifier type="IEC" primary="true">IEC 60050-103:2009</docidentifier>
           <docidentifier type='URN'>urn:iec:std:iec:60050-103:2009:::en</docidentifier>
           <date type="published">
             <on>2009-12-14</on>
@@ -1288,7 +1288,7 @@ RSpec.describe Metanorma::Standoc do
              <sourcecode id='L__xf6_we'>
                <name>
                  See
-                 <eref type='inline' bibitemid='L__xf6_wner2016' citeas='L&#246;wner et al. 2016'/>
+                 <eref type='inline' bibitemid='L__xf6_wner2016' citeas='L&amp;#xf6;wner et al. 2016'/>
                </name>
                ABC
              </sourcecode>

data/spec/metanorma/isobib_cache_spec.rb

@@ -49,7 +49,7 @@ ISO_124_DATED = <<~XML.freeze
          <uri type="src">https://www.iso.org/standard/61884.html</uri>
          <uri type="obp">https://www.iso.org/obp/ui/#!iso:std:61884:en</uri>
          <uri type="rss">https://www.iso.org/contents/data/standard/06/18/61884.detail.rss</uri>
-         <docidentifier type="ISO">ISO 124:2014</docidentifier>
+         <docidentifier type="ISO" primary="true">ISO 124:2014</docidentifier>
          <docidentifier type='URN'>urn:iso:std:iso:124:stage-90.93:ed-7:en</docidentifier>
          <docnumber>124</docnumber>
          <date type="published">
@@ -208,7 +208,7 @@ ISO_123_DATED = <<~XML.freeze
            <uri type="src">https://www.iso.org/standard/23281.html</uri>
            <uri type="obp">https://www.iso.org/obp/ui/#!iso:std:23281:en</uri>
            <uri type="rss">https://www.iso.org/contents/data/standard/02/32/23281.detail.rss</uri>
-           <docidentifier type="ISO">ISO 123:2001</docidentifier>
+           <docidentifier type="ISO" primary="true">ISO 123:2001</docidentifier>
            <docidentifier type='URN'>urn:iso:std:iso:123:stage-90.93:ed-3:en</docidentifier>
            <docnumber>123</docnumber>
            <date type="published">