metanorma-standoc 1.7.2 → 1.8.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: b74869d08fd6dece972c8a91e6d3bee248f3cd1f8f6dbf5ce1280358be4a6e27
-  data.tar.gz: d3d41f3fc77fead3848808777fcfaf09c20baa24a3aebdde2ace278fcbb44cfd
+  metadata.gz: d069ef191d60dfccff15d8a39e2c961c343f1e19928e06645c78833be951f052
+  data.tar.gz: 61dff9a129459b4f2b0c4cb426dabd84a822285c8cce41cd88ce20cc4706d3a5
 SHA512:
-  metadata.gz: da9c7ea38fa9545e6657d178296a7c022a4ab460ba8ea31a9e34b188e7be5451c2ff6e4e202767f4ca57a1548803744f7bdec6bac1b1a1684429fe2b66be1118
-  data.tar.gz: 2e67fb7b6f31ca587c37855d65ede4ef7ea3c9153eedce5a00f7003312b20cb87821060ecf4abdc3637a6aaa5899847659c153491e18a1e2fce81d7dd320bd8b
+  metadata.gz: 9a5991f1f7b0262f7b2e3069572d4221f0f944be86a869e9a6bb5bdabe26441305882898788840ef373dfb1671d7c2a8d9d59e1d5942e8d2df6ccc17eee1c114
+  data.tar.gz: 7e9c4dbf9f8d28970756ed8d516df46d5a4c2710754065818c2966de478ed32e1d189d0eec12de1ca5b0b23de5e077f9c4254e738442a7d4a1005e62d6a304d1

data/.github/workflows/rake.yml

@@ -36,6 +36,9 @@ jobs:
         with:
           ruby-version: ${{ matrix.ruby }}
 
+      - if: matrix.os == 'macos-latest'
+        run: brew install autoconf automake libtool
+
       - uses: actions/cache@v2
         with:
           path: vendor/bundle
@@ -64,3 +67,17 @@ jobs:
         run: cinst -y plantuml
 
       - run: bundle exec rake
+
+  tests-passed:
+    needs: rake
+    runs-on: ubuntu-latest
+    continue-on-error: true
+    steps:
+      - name: Trigger tests passed event
+        uses: Sibz/github-status-action@v1
+        with:
+          authToken: ${{ secrets.METANORMA_CI_PAT_TOKEN || secrets.GITHUB_TOKEN }}
+          context: 'tests-passed-successfully'
+          description: 'Tests passed successfully'
+          state: 'success'
+          sha: ${{ github.event.pull_request.head.sha || github.sha }}

data/.rubocop.yml

@@ -5,6 +5,6 @@
 inherit_from:
   - https://raw.githubusercontent.com/riboseinc/oss-guides/master/ci/rubocop.yml
 AllCops:
-  TargetRubyVersion: 2.6
+  TargetRubyVersion: 2.4
 Rails:
   Enabled: true

data/lib/asciidoctor/standoc/base.rb

@@ -7,6 +7,7 @@ require "open-uri"
 require "isodoc"
 require "relaton"
 require "fileutils"
+require "metanorma-utils"
 
 module Asciidoctor
   module Standoc
@@ -91,6 +92,7 @@ module Asciidoctor
         @draft = false
         @refids = Set.new
         @anchors = {}
+        @internal_eref_namespaces = []
         @draft = node.attributes.has_key?("draft")
         @novalid = node.attr("novalid")
         @smartquotes = node.attr("smartquotes") != "false"
@@ -98,28 +100,32 @@ module Asciidoctor
         @fontheader = default_fonts(node)
         @files_to_delete = []
         @filename = node.attr("docfile") ?  File.basename(node.attr("docfile")).gsub(/\.adoc$/, "") : ""
-        @localdir = Utils::localdir(node)
+        @localdir = Metanorma::Utils::localdir(node)
         @output_dir = outputdir node
         @no_isobib_cache = node.attr("no-isobib-cache")
         @no_isobib = node.attr("no-isobib")
+        @sourcecode_markup_start = node.attr("sourcecode-markup-start") || "{{{"
+        @sourcecode_markup_end = node.attr("sourcecode-markup-end") || "}}}"
         @bibdb = nil
         @seen_headers = []
         @datauriimage = node.attr("data-uri-image")
         @boilerplateauthority = node.attr("boilerplate-authority")
-        @log = Asciidoctor::Standoc::Log.new
+        @sourcecode_markup_start = node.attr("sourcecode-markup-start") || "{{{"
+        @sourcecode_markup_end = node.attr("sourcecode-markup-end") || "}}}"
+        @log = Metanorma::Utils::Log.new
         init_bib_caches(node)
         init_iev_caches(node)
         @lang = (node.attr("language") || "en")
-        @script = (node.attr("script") || "Latn")
+        @script = (node.attr("script") || default_script(node.attr("language")))
         @isodoc = isodoc(@lang, @script, node.attr("i18nyaml"))
         @i18n = @isodoc.i18n
       end
 
       def default_fonts(node)
         b = node.attr("body-font") ||
-          (node.attr("script") == "Hans" ? '"SimSun",serif' : '"Cambria",serif')
+          (node.attr("script") == "Hans" ? '"Source Han Sans",serif' : '"Cambria",serif')
         h = node.attr("header-font") ||
-          (node.attr("script") == "Hans" ? '"SimHei",sans-serif' : '"Cambria",serif')
+          (node.attr("script") == "Hans" ? '"Source Han Sans",sans-serif' : '"Cambria",serif')
         m = node.attr("monospace-font") || '"Courier New",monospace'
         "$bodyfont: #{b};\n$headerfont: #{h};\n$monospacefont: #{m};\n"
       end
@@ -133,7 +139,7 @@ module Asciidoctor
 
       def document(node)
         init(node)
-        ret = makexml(node).to_xml(indent: 2)
+        ret = makexml(node).to_xml(encoding: "US-ASCII", indent: 2)
         outputs(node, ret) unless node.attr("nodoc") || !node.attr("docfile")
         clean_exit
         ret
@@ -185,52 +191,35 @@ module Asciidoctor
         end
       end
 
-      def term_source_attrs(seen_xref)
-        { bibitemid: seen_xref.children[0]["target"],
-          format: seen_xref.children[0]["format"], type: "inline" }
-      end
-
-      def add_term_source(xml_t, seen_xref, m)
-        if seen_xref.children[0].name == "concept"
-          xml_t.origin { |o| o << seen_xref.children[0].to_xml }
+      def default_script(lang)
+        case lang
+        when "ar", "fa"
+          "Arab"
+        when "ur"
+          "Aran"
+        when "ru", "bg"
+          "Cyrl"
+        when "hi"
+          "Deva"
+        when "el"
+          "Grek"
+        when "zh"
+          "Hans"
+        when "ko"
+          "Kore"
+        when "he"
+          "Hebr"
+        when "ja"
+          "Jpan"
         else
-          xml_t.origin seen_xref.children[0].content, **attr_code(term_source_attrs(seen_xref))
+          "Latn"
         end
-        m[:text] && xml_t.modification do |mod|
-          mod.p { |p| p << m[:text].sub(/^\s+/, "") }
-        end
-      end
-
-      TERM_REFERENCE_RE_STR = <<~REGEXP.freeze
-        ^(?<xref><(xref|concept)[^>]+>([^<]*</(xref|concept)>)?)
-               (,\s(?<text>.*))?
-        $
-      REGEXP
-      TERM_REFERENCE_RE =
-        Regexp.new(TERM_REFERENCE_RE_STR.gsub(/\s/, "").gsub(/_/, "\\s"),
-                   Regexp::IGNORECASE | Regexp::MULTILINE)
-
-      def extract_termsource_refs(text, node)
-        matched = TERM_REFERENCE_RE.match text
-        matched.nil? and @log.add("AsciiDoc Input", node, "term reference not in expected format: #{text}")
-        matched
-      end
-
-      def termsource(node)
-        matched = extract_termsource_refs(node.content, node) || return
-        noko do |xml|
-          attrs = { status: matched[:text] ? "modified" : "identical" }
-          xml.termsource **attrs do |xml_t|
-            seen_xref = Nokogiri::XML.fragment(matched[:xref])
-            add_term_source(xml_t, seen_xref, matched)
-          end
-        end.join("\n")
       end
 
       private
 
       def outputdir(node)
-        if node.attr("output_dir").nil_or_empty? then Utils::localdir(node)
+        if node.attr("output_dir").nil_or_empty? then Metanorma::Utils::localdir(node)
         else File.join(node.attr("output_dir"), "")
         end
       end

data/lib/asciidoctor/standoc/basicdoc.rng

@@ -729,9 +729,11 @@
   </define>
   <define name="index">
     <element name="index">
-      <attribute name="to">
-        <data type="IDREF"/>
-      </attribute>
+      <optional>
+        <attribute name="to">
+          <data type="IDREF"/>
+        </attribute>
+      </optional>
       <element name="primary">
         <oneOrMore>
           <ref name="PureTextElement"/>

data/lib/asciidoctor/standoc/blocks.rb

@@ -6,11 +6,11 @@ module Asciidoctor
   module Standoc
     module Blocks
       def id_attr(node = nil)
-        { id: Utils::anchor_or_uuid(node) }
+        { id: Metanorma::Utils::anchor_or_uuid(node) }
       end
 
       def id_unnum_attrs(node)
-        attr_code( id: Utils::anchor_or_uuid(node),
+        attr_code( id: Metanorma::Utils::anchor_or_uuid(node),
                   unnumbered: node.option?("unnumbered") ? "true" : nil,
                   number: node.attr("number"),
                   subsequence: node.attr("subsequence") )
@@ -29,7 +29,7 @@ module Asciidoctor
       # We append each contained block to its parent
       def open(node)
         role = node.role || node.attr("style")
-        Utils::reqt_subpart(role) and return requirement_subpart(node)
+        reqt_subpart(role) and return requirement_subpart(node)
         result = []
         node.blocks.each do |b|
           result << send(b.context, b)
@@ -45,7 +45,7 @@ module Asciidoctor
         noko do |xml|
           xml.figure **literal_attrs(node) do |f|
             figure_title(node, f)
-            f.pre node.lines.join("\n"), **attr_code(id: Utils::anchor_or_uuid,
+            f.pre node.lines.join("\n"), **attr_code(id: Metanorma::Utils::anchor_or_uuid,
                                                      alt: node.attr("alt"))
           end
         end
@@ -74,11 +74,20 @@ module Asciidoctor
         %w(recommendation requirement permission).include?(role) and
           return requirement(node, role)
         return pseudocode_example(node) if role == "pseudocode"
+        return svgmap_example(node) if role == "svgmap"
         example_proper(node)
       end
 
+      def svgmap_example(node)
+        noko do |xml|
+          xml.svgmap **attr_code(example_attrs(node).merge(src: node.attr("src"), alt: node.attr("alt"))) do |ex|
+            ex << node.content
+          end
+        end.join("\n")
+      end
+
+      # prevent A's and other subs inappropriate for pseudocode
       def pseudocode_example(node)
-        # prevent A's and other subs inappropriate for pseudocode
         node.blocks.each { |b| b.remove_sub(:replacements) }
         noko do |xml|
           xml.figure **example_attrs(node).merge(class: "pseudocode") do |ex|
@@ -121,7 +130,7 @@ module Asciidoctor
 
       def para_attrs(node)
         attr_code(keep_attrs(node).merge(align: node.attr("align"), 
-                                        id: Utils::anchor_or_uuid(node)))
+                                         id: Metanorma::Utils::anchor_or_uuid(node)))
       end
 
       def paragraph(node)
@@ -135,7 +144,7 @@ module Asciidoctor
 
       def quote_attrs(node)
         attr_code(keep_attrs(node).merge(align: node.attr("align"), 
-                                        id: Utils::anchor_or_uuid(node)))
+                                         id: Metanorma::Utils::anchor_or_uuid(node)))
       end
 
       def quote_attribution(node, out)
@@ -160,10 +169,10 @@ module Asciidoctor
 
       def listing_attrs(node)
         attr_code(keep_attrs(node).merge(lang: node.attr("language"),
-                                        id: Utils::anchor_or_uuid(node),
-                                        unnumbered: node.option?("unnumbered") ? "true" : nil,
-                                        number: node.attr("number"),
-                                        filename: node.attr("filename")))
+                                         id: Metanorma::Utils::anchor_or_uuid(node),
+                                         unnumbered: node.option?("unnumbered") ? "true" : nil,
+                                         number: node.attr("number"),
+                                         filename: node.attr("filename")))
       end
 
       # NOTE: html escaping is performed by Nokogiri

data/lib/asciidoctor/standoc/blocks_notes.rb

@@ -35,7 +35,7 @@ module Asciidoctor
         date = node.attr("date") || Date.today.iso8601.gsub(/\+.*$/, "")
         date += "T00:00:00Z" unless /T/.match date
         attr_code(
-          id: Utils::anchor_or_uuid(node),
+          id: ::Metanorma::Utils::anchor_or_uuid(node),
           reviewer: node.attr("reviewer") || node.attr("source") || "(Unknown)",
           date: date )
       end
@@ -69,7 +69,7 @@ module Asciidoctor
         a = node.attr("type") and ["danger", "safety precautions"].each do |t|
           name = t if a.casecmp(t).zero?
         end
-        attr_code(keep_attrs(node).merge(id: Utils::anchor_or_uuid(node), type: name,
+        attr_code(keep_attrs(node).merge(id: Metanorma::Utils::anchor_or_uuid(node), type: name,
                   beforeclauses: node.attr("beforeclauses") == "true" ? "true" : nil))
       end
 

data/lib/asciidoctor/standoc/cleanup.rb

@@ -5,11 +5,13 @@ require "html2doc"
 require_relative "./cleanup_block.rb"
 require_relative "./cleanup_footnotes.rb"
 require_relative "./cleanup_ref.rb"
+require_relative "./cleanup_ref_dl.rb"
 require_relative "./cleanup_boilerplate.rb"
 require_relative "./cleanup_section.rb"
 require_relative "./cleanup_terms.rb"
 require_relative "./cleanup_inline.rb"
 require_relative "./cleanup_amend.rb"
+require_relative "./cleanup_maths.rb"
 require "relaton_iev"
 
 module Asciidoctor
@@ -23,26 +25,15 @@ module Asciidoctor
                   </passthrough>}mx) { |m| HTMLEntities.new.decode($1) }
       end
 
-      def asciimath2mathml(text)
-        text = text.gsub(%r{<stem type="AsciiMath">(.+?)</stem>}m) do |m|
-            "<amathstem>#{HTMLEntities.new.decode($1)}</amathstem>"
-          end
-          text = Html2Doc.
-            asciimath_to_mathml(text, ["<amathstem>", "</amathstem>"])
-          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
-      end
-
       def cleanup(xmldoc)
         element_name_cleanup(xmldoc)
         sections_cleanup(xmldoc)
         obligations_cleanup(xmldoc)
         table_cleanup(xmldoc)
         formula_cleanup(xmldoc)
+        sourcecode_cleanup(xmldoc)
         figure_cleanup(xmldoc)
+        element_name_cleanup(xmldoc)
         ref_cleanup(xmldoc)
         note_cleanup(xmldoc)
         clausebefore_cleanup(xmldoc)
@@ -67,6 +58,7 @@ module Asciidoctor
         bookmark_cleanup(xmldoc)
         requirement_cleanup(xmldoc)
         bibdata_cleanup(xmldoc)
+        svgmap_cleanup(xmldoc)
         boilerplate_cleanup(xmldoc)
         smartquotes_cleanup(xmldoc)
         variant_cleanup(xmldoc)
@@ -78,14 +70,13 @@ module Asciidoctor
       end
 
       def smartquotes_cleanup(xmldoc)
-        xmldoc.xpath("//date").each { |d| Utils::endash_date(d) }
+        xmldoc.xpath("//date").each { |d| Metanorma::Utils::endash_date(d) }
         xmldoc.traverse do |n|
           next unless n.text?
           if @smartquotes
             /[-'"(<>]|\.\.|\dx/.match(n) or next
-            n.ancestors("pre, tt, sourcecode, bibdata, on, "\
-                        "stem, figure[@class = 'pseudocode']").empty? or next
-            n.replace(Utils::smartformat(n.text))
+            n.ancestors("pre, tt, sourcecode, bibdata, on, stem, figure[@class = 'pseudocode']").empty? or next
+            n.replace(Metanorma::Utils::smartformat(n.text))
           else
             n.replace(n.text.gsub(/(?<=\p{Alnum})\u2019(?=\p{Alpha})/, "'"))#.
             #gsub(/</, "&lt;").gsub(/>/, "&gt;"))
@@ -148,60 +139,26 @@ module Asciidoctor
         align_callouts_to_annotations(xmldoc)
       end
 
-      def xml_unescape_mathml(x)
-        return if x.children.any? { |y| y.element? }
-        math = x.text.gsub(/&lt;/, "<").gsub(/&gt;/, ">").gsub(/&quot;/, '"').
-          gsub(/&apos;/, "'").gsub(/&amp;/, "&").
-          gsub(/<[^: \r\n\t\/]+:/, "<").gsub(/<\/[^ \r\n\t:]+:/, "</")
-        x.children = math
-      end
-
-      MATHML_NS = "http://www.w3.org/1998/Math/MathML".freeze
-
-      def mathml_preserve_space(m)
-        m.xpath(".//m:mtext", "m" => MATHML_NS).each do |x|
-          x.children = x.children.to_xml.gsub(/^\s/, "&#xA0;").
-            gsub(/\s$/, "&#xA0;")
-        end
-      end
-
-      def mathml_namespace(stem)
-        stem.xpath("./math", ).each { |x| x.default_namespace = MATHML_NS }
-      end
-
-      def mathml_mi_italics
-        { uppergreek: true, upperroman: true,
-          lowergreek: true, lowerroman: true }
-      end
-
-      # presuppose multichar mi upright, singlechar mi MathML default italic
-      def mathml_italicise(x)
-        x.xpath(".//m:mi[not(ancestor::*[@mathvariant])]", "m" => MATHML_NS).each do |i|
-          char = HTMLEntities.new.decode(i.text)
-          i["mathvariant"] = "normal" if mi_italicise?(char)
-        end
-      end
-
-      def mi_italicise?(c)
-        return false if c.length > 1
-        if /\p{Greek}/.match(c)
-          /\p{Lower}/.match(c) && !mathml_mi_italics[:lowergreek] ||
-            /\p{Upper}/.match(c) && !mathml_mi_italics[:uppergreek]
-        elsif /\p{Latin}/.match(c)
-          /\p{Lower}/.match(c) && !mathml_mi_italics[:lowerroman] ||
-            /\p{Upper}/.match(c) && !mathml_mi_italics[:upperroman]
-        else
-          false
+      def sourcecode_cleanup(xmldoc)
+        xmldoc.xpath("//sourcecode").each do |x|
+          x.traverse do |n|
+            next unless n.text?
+            next unless /#{Regexp.escape(@sourcecode_markup_start)}/.match(n.text)
+            n.replace(sourcecode_markup(n))
+          end
         end
       end
 
-      def mathml_cleanup(xmldoc)
-        xmldoc.xpath("//stem[@type = 'MathML']").each do |x|
-          xml_unescape_mathml(x)
-          mathml_namespace(x)
-          mathml_preserve_space(x)
-          mathml_italicise(x)
+      def sourcecode_markup(n)
+        acc = []
+        n.text.split(/(#{Regexp.escape(@sourcecode_markup_start)}|#{Regexp.escape(@sourcecode_markup_end)})/).
+          each_slice(4).map do |a|
+          acc << Nokogiri::XML::Text.new(a[0], n.document).
+            to_xml(encoding: "US-ASCII", save_with: Nokogiri::XML::Node::SaveOptions::NO_DECLARATION)
+          next unless a.size == 4
+          acc << Asciidoctor.convert(a[2], backend: (self&.backend&.to_sym || :standoc), doctype: :inline)
         end
+        acc.join
       end
 
       # allows us to deal with doc relation localities,
@@ -215,23 +172,18 @@ module Asciidoctor
 
       def img_cleanup(xmldoc)
         return xmldoc unless @datauriimage
-        xmldoc.xpath("//image").each do |i|
-          i["src"] = datauri(i["src"])
-        end
+        xmldoc.xpath("//image").each { |i| i["src"] = Metanorma::Utils::datauri(i["src"], @localdir) }
       end
 
       def variant_cleanup(xmldoc)
         xmldoc.xpath("//*[variant]").each do |c|
-          c&.next&.text? && c&.next&.next&.name == "variant" &&
-            c.next.text.gsub(/\s/, "").empty? and c.next.remove
+          c&.next&.text? && c&.next&.next&.name == "variant" && c.next.text.gsub(/\s/, "").empty? and
+            c.next.remove
         end
         xmldoc.xpath("//*[variant]").each do |c|
-          next unless c.children.any? do |n|
-            n.name != "variant" && (!n.text? || !n.text.gsub(/\s/, "").empty?)
-          end
+          next unless c.children.any? { |n| n.name != "variant" && (!n.text? || !n.text.gsub(/\s/, "").empty?) }
           c.xpath("./variant").each do |n|
-            if n.at_xpath('preceding-sibling::node()[not(self::text()'\
-                '[not(normalize-space())])][1][self::variantwrap]')
+            if n.at_xpath('preceding-sibling::node()[not(self::text()[not(normalize-space())])][1][self::variantwrap]')
               n.previous_element << n
             else
               n.replace('<variantwrap/>').first << n