metanorma-ieee 0.0.1 → 0.0.4

data/lib/isodoc/ieee/word_authority.rb

@@ -46,6 +46,7 @@ module IsoDoc
           officeholder_style(p)
         end
         officemember_style(docxml)
+        officeorgrep_style(docxml)
         three_column_officemembers(docxml
           .at("//div[@id = 'boilerplate-participants']"))
       end
@@ -53,9 +54,9 @@ module IsoDoc
       def officeholder_style(para)
         n = para.next_element
         p = para.previous_element
-        n && n.name == "p" && n["type"] == "officeholder" and
+        n && n.name == "p" && n["type"] != "officeholder" and
           klass = "IEEEStdsNamesCtrCxSpLast"
-        p && p.name == "p" && p["type"] == "officeholder" and
+        p && p.name == "p" && p["type"] != "officeholder" and
           klass = "IEEEStdsNamesCtrCxSpFirst"
         para["class"] = klass || "IEEEStdsNamesCtrCxSpMiddle"
       end
@@ -68,7 +69,8 @@ module IsoDoc
       BREAK
 
       def officemember_style(docxml)
-        docxml.xpath("//p[@type = 'officemember']").each do |p|
+        docxml.xpath("//p[@type = 'officemember' or @type = 'officeorgmember']")
+          .each do |p|
           p["class"] = "IEEEStdsNamesList"
         end
         docxml.xpath("//p[@type = 'emeritus_sign']").each do |p|
@@ -76,6 +78,19 @@ module IsoDoc
         end
       end
 
+      def officeorgrep_style(docxml)
+        docxml.xpath("//p[@type = 'officeorgrepmemberhdr']").each do |p|
+          p["class"] = "IEEEStdsNamesList"
+          p["style"] =
+            "margin-bottom:6.0pt;tab-stops:right 432.0pt;"
+        end
+        docxml.xpath("//p[@type = 'officeorgrepmember']").each do |p|
+          p["class"] = "IEEEStdsNamesList"
+          p["style"] =
+            "margin-top:6.0pt;tab-stops:right dotted 432.0pt;"
+        end
+      end
+
       def three_column_officemembers(div)
         return unless div
 
@@ -85,6 +100,7 @@ module IsoDoc
 
       def three_column_officemembers_split(div)
         prev = false
+        div.xpath(".//div").each { |d| d.replace(d.children) }
         div.elements.each_with_object([[]]) do |e, m|
           member = e.name == "p" && e["type"] == "officemember"
           (prev == member and m[-1] << e.to_xml) or m << [e.to_xml]
@@ -125,6 +141,10 @@ module IsoDoc
         docxml.at("//div[@class = 'boilerplate-feedback']")&.xpath("./div")
           &.each_with_index do |div, i|
           i.zero? or div.elements.first.previous = "<p>&#xa0;</p>"
+          i == 4 and
+            div.xpath(".//p[br]").each do |p|
+              p.replace(p.to_xml.gsub(%r{<br/>}, "</p><p>"))
+            end
           feedback_style1(div, i)
         end
       end
@@ -155,6 +175,56 @@ module IsoDoc
           end
         end
       end
+
+      def abstract_cleanup(docxml)
+        dest = docxml.at("div[@id = 'abstract-destination']") or return
+        if f = docxml.at("//div[@class = 'abstract']")
+          f.previous_element.remove
+          abstract_cleanup1(f, dest)
+          f.remove
+        elsif f = docxml.at("//div[@type = 'scope']")
+          abstract_cleanup1(f, dest)
+          abstract_header(dest)
+        end
+      end
+
+      def abstract_cleanup1(source, dest)
+        source.elements.reject { |e| %w(h1 h2).include?(e.name) }.each do |e|
+          e1 = e.dup
+          e1.xpath(".//p").each do |p|
+            p["style"] ||= ""
+            p["style"] = 'font-family: "Arial", sans-serif;' + p["style"]
+          end
+          %w(ul ol).include?(e1.name) or e1["class"] = "IEEEStdsAbstractBody"
+          dest << e1
+        end
+      end
+
+      def abstract_header(dest)
+        dest.elements.first.children.first.previous =
+          "<span class='IEEEStdsAbstractHeader'><span lang='EN-US'>"\
+          "Abstract:</span></span> "
+      end
+
+      def introduction_cleanup(docxml)
+        dest = docxml.at("div[@id = 'introduction-destination']") or return
+        unless i = docxml.at("//h1[@class = 'IntroTitle']")&.parent
+          dest.parent.remove
+          return
+        end
+        introduction_cleanup1(i, dest)
+      end
+
+      def introduction_cleanup1(intro, dest)
+        docxml = intro.document
+        intro.previous_element.remove
+        dest.replace(intro.remove)
+        i = docxml.at("//h1[@class = 'IntroTitle']")
+        if i.next_element.name == "div" &&
+            i.next_element["class"] == "IEEEStdsIntroduction"
+          i.next_element.name = "p"
+        end
+      end
     end
   end
 end

data/lib/isodoc/ieee/word_cleanup.rb

@@ -17,70 +17,65 @@ module IsoDoc
         @wordstylesheet.unlink if @wordstylesheet.is_a?(Tempfile)
       end
 
-      def abstract_cleanup(docxml)
-        dest = docxml.at("div[@id = 'abstract-destination']") or return
-        if f = docxml.at("//div[@class = 'abstract']")
-          f.previous_element.remove
-          abstract_cleanup1(f, dest)
-          f.remove
-        elsif f = docxml.at("//div[@type = 'scope']")
-          abstract_cleanup1(f, dest)
-        end
-      end
-
-      def abstract_cleanup1(source, dest)
-        source.elements.each do |e|
-          next if %w(h1 h2).include?(e.name)
-
-          dest << e.dup
-          dest.elements.last["class"] = "IEEEStdsAbstractBody"
-        end
-        dest.elements.first.children.first.previous =
-          "<span class='IEEEStdsAbstractHeader'><span lang='EN-US'>"\
-          "Abstract:</span></span> "
-      end
-
-      def introduction_cleanup(docxml)
-        dest = docxml.at("div[@id = 'introduction-destination']") or return
-        unless i = docxml.at("//h1[@class = 'IntroTitle']")&.parent
-          dest.parent.remove
-          return
-        end
-        introduction_cleanup1(i, dest)
-      end
-
-      def introduction_cleanup1(intro, dest)
-        docxml = intro.document
-        intro.previous_element.remove
-        dest.replace(intro.remove)
-        i = docxml.at("//h1[@class = 'IntroTitle']")
-        i.next_element == "div" && i.next_element["class"] == "Admonition" and
-          i.next_element["class"] = "IEEEStdsIntroduction"
-      end
-
       def word_cleanup(docxml)
         super
         abstract_cleanup(docxml)
         introduction_cleanup(docxml)
+        sourcecode_cleanup(docxml)
         div_cleanup(docxml)
+        biblio_cleanup(docxml)
         headings_cleanup(docxml)
-        span_style_cleanup(docxml)
+        caption_cleanup(docxml)
+        table_cleanup(docxml)
         style_cleanup(docxml)
         para_type_cleanup(docxml)
         docxml
       end
 
+      def make_WordToC(docxml, level)
+        toc = ""
+        xpath = (1..level).each.map do |i|
+          "//h#{i}[not(ancestor::*[@class = 'WordSection2'])]"
+        end.join (" | ")
+        docxml.xpath(xpath).each do |h|
+          toc += word_toc_entry(h.name[1].to_i, header_strip(h))
+        end
+        toc.sub(/(<p class="MsoToc1">)/,
+                %{\\1#{word_toc_preface(level)}}) + WORD_TOC_SUFFIX1
+      end
+
+      def biblio_cleanup(docxml)
+        docxml.xpath("//p[@class = 'Biblio']").each do |p|
+          headings_strip(p)
+        end
+      end
+
       def headings_cleanup(docxml)
-        (1..4).each do |i|
-          docxml.xpath("//h#{i}").each do |h|
-            headings_cleanup1(h)
-            h.name = "p"
-            h["class"] = "IEEEStdsLevel#{i}Header"
-          end
+        (1..9).each { |i| headings_cleanup1(docxml, i) }
+        docxml.xpath("//div[@class = 'Annex']").each { |a| a.delete("class") }
+      end
+
+      def headings_cleanup1(docxml, idx)
+        docxml.xpath("//h#{idx}").each do |h|
+          headings_strip(h)
+          headings_style(h, idx)
+        end
+      end
+
+      def headings_style(hdr, idx)
+        if hdr.at("./ancestor::div[@class = 'Annex']")
+          hdr.delete("class")
+          hdr["style"] = "mso-list:l13 level#{idx} lfo33;"
+        elsif hdr.at("./ancestor::div[@class = 'Section3']")
+          hdr.name = "p"
+          hdr["class"] = "IEEEStdsLevel#{idx}frontmatter"
+        else
+          hdr.name = "p"
+          hdr["class"] = "IEEEStdsLevel#{idx}Header"
         end
       end
 
-      def headings_cleanup1(hdr)
+      def headings_strip(hdr)
         if hdr.children.size > 1 && hdr.children[1].name == "span" &&
             hdr.children[1]["style"] == "mso-tab-count:1"
           2.times { hdr.children.first.remove }
@@ -100,28 +95,23 @@ module IsoDoc
         end
       end
 
-      def para_type_cleanup(html)
-        html.xpath("//p[@type]").each { |p| p.delete("type") }
-      end
-
-      def span_style_cleanup(html)
-        html.xpath("//strong").each do |s|
-          s.name = "span"
-          s["class"] = "IEEEStdsParaBold"
-        end
-        html.xpath("//em").each do |s|
-          s.name = "span"
-          s["class"] = "IEEEStdsAddItal"
-        end
-      end
-
       STYLESMAP = {
+        example: "IEEEStdsParagraph",
         MsoNormal: "IEEEStdsParagraph",
-        NormRef: "IEEEStdsBibliographicEntry",
+        NormRef: "IEEEStdsParagraph",
         Biblio: "IEEEStdsBibliographicEntry",
+        figure: "IEEEStdsImage",
+        formula: "IEEEStdsEquation",
+        Sourcecode: "IEEEStdsComputerCode",
+        TableTitle: "IEEEStdsRegularTableCaption",
+        FigureTitle: "IEEEStdsRegularFigureCaption",
       }.freeze
 
       def style_cleanup(docxml)
+        note_style_cleanup(docxml)
+        docxml.xpath("//div[@class = 'formula']/p").each do |p|
+          p["class"] = "IEEEStdsEquation"
+        end
         STYLESMAP.each do |k, v|
           docxml.xpath("//*[@class = '#{k}']").each { |s| s["class"] = v }
         end

data/lib/isodoc/ieee/word_cleanup_blocks.rb

@@ -0,0 +1,151 @@
+module IsoDoc
+  module IEEE
+    class WordConvert < IsoDoc::WordConvert
+      def admonition_cleanup(docxml)
+        super
+        docxml.xpath("//div[@class = 'zzHelp']").each do |d|
+          d.xpath(".//p").each do |p|
+            %w(IEEEStdsWarning IEEEStdsParagraph).include?(p["class"]) ||
+              !p["class"] or next
+
+            p["class"] = "zzHelp"
+          end
+        end
+        docxml
+      end
+
+      def table_cleanup(docxml)
+        thead_cleanup(docxml)
+        tbody_cleanup(docxml)
+      end
+
+      def thead_cleanup(docxml)
+        docxml.xpath("//thead").each do |h|
+          h.xpath(".//td | .//th").each do |t|
+            if t.at("./p")
+              t.xpath("./p").each do |p|
+                p["class"] = "IEEEStdsTableColumnHead"
+              end
+            else
+              t.children =
+                "<p class='IEEEStdsTableColumnHead'>#{t.children.to_xml}</p>"
+            end
+          end
+        end
+      end
+
+      def tbody_cleanup(docxml)
+        docxml.xpath("//tbody | //tfoot").each do |h|
+          next if h.at("./ancestor::div[@class = 'boilerplate-feedback']")
+
+          h.xpath(".//th").each { |t| tbody_head_cleanup(t) }
+          h.xpath(".//td | .//th").each { |t| tbody_cleanup1(t) }
+        end
+      end
+
+      def tbody_head_cleanup(cell)
+        cell.at("./p") or
+          cell.children = "<p>#{cell.children.to_xml}</p>"
+        cell.xpath("./p").each do |p|
+          p.replace p.to_xml.gsub(%r{<br/>}, "</p><p>")
+        end
+      end
+
+      def tbody_cleanup1(cell)
+        if cell.at("./p")
+          cell.xpath("./p").each_with_index do |p, i|
+            p["class"] = td_style(cell, i)
+          end
+        else
+          cell.children =
+            "<p class='#{td_style(cell, 0)}'>#{cell.children.to_xml}</p>"
+        end
+      end
+
+      def td_style(cell, idx)
+        if cell.name == "th" && idx.zero? then "IEEEStdsTableLineHead"
+        elsif cell.name == "th" then "IEEEStdsTableLineSubhead"
+        elsif cell["align"] == "center" ||
+            /text-align:center/.match?(cell["style"])
+          "IEEEStdsTableData-Center"
+        else "IEEEStdsTableData-Left"
+        end
+      end
+
+      def caption_cleanup(docxml)
+        table_caption(docxml)
+        figure_caption(docxml)
+        example_caption(docxml)
+      end
+
+      def table_caption(docxml)
+        docxml.xpath("//p[@class = 'TableTitle']").each do |s|
+          s.children = s.children.to_xml
+            .sub(/^#{@i18n.table}(\s+[A-Z0-9.]+)?/, "")
+        end
+      end
+
+      def figure_caption(docxml)
+        docxml.xpath("//p[@class = 'FigureTitle']").each do |s|
+          s.children = s.children.to_xml
+            .sub(/^#{@i18n.figure}(\s+[A-Z0-9.]+)?/, "")
+        end
+      end
+
+      def example_caption(docxml)
+        docxml.xpath("//p[@class = 'example-title']").each do |s|
+          s.children = "<em>#{s.children.to_xml}</em>"
+          s["class"] = "IEEEStdsParagraph"
+        end
+      end
+
+      def sourcecode_cleanup(docxml)
+        docxml.xpath("//p[@class = 'Sourcecode']").each do |s|
+          s.replace(s.to_xml.gsub(%r{<br/>}, "</p><p class='Sourcecode'>"))
+        end
+      end
+
+      def para_type_cleanup(html)
+        html.xpath("//p[@type]").each { |p| p.delete("type") }
+      end
+
+      def note_style_cleanup(docxml)
+        docxml.xpath("//span[@class = 'note_label']").each do |s|
+          multi = /^#{@i18n.note}\s+[A-Z0-9.]+/.match?(s.text)
+          div = s.at("./ancestor::div[@class = 'Note']")
+          if multi
+            s.remove
+            seq = notesequence(div)
+          else seq = nil
+          end
+          note_style_cleanup1(multi, div, seq)
+        end
+      end
+
+      def notesequence(div)
+        @notesequences ||= { max: 0, lookup: {} }
+        unless id = @notesequences[:lookup][@xrefs.anchor(div["id"],
+                                                          :sequence)]
+          @notesequences[:max] += 1
+          id = @notesequences[:max]
+          @notesequences[:lookup][@xrefs.anchor(div["id"], :sequence)] =
+            id
+        end
+        id
+      end
+
+      # hardcoded list style for notes
+      def note_style_cleanup1(multi, div, seq)
+        div.xpath(".//p[@class = 'Note' or not(@class)]")
+          .each_with_index do |p, i|
+          p["class"] =
+            i.zero? && multi ? "IEEEStdsMultipleNotes" : "IEEEStdsSingleNote"
+          if multi
+            p["style"] ||= ""
+            p["style"] += "mso-list:l17 level1 lfo#{seq};"
+          end
+        end
+      end
+    end
+  end
+end

data/lib/isodoc/ieee/word_convert.rb

@@ -1,6 +1,7 @@
 require "isodoc"
 require_relative "init"
 require_relative "word_cleanup"
+require_relative "word_cleanup_blocks"
 require_relative "word_authority"
 
 module IsoDoc
@@ -11,6 +12,14 @@ module IsoDoc
         super
       end
 
+      def convert1(docxml, filename, dir)
+        doctype = docxml.at(ns("//bibdata/ext/doctype"))
+        if %w(amendment corrigendum).include?(doctype&.text)
+          @header = html_doc_path("header_amd.html")
+        end
+        super
+      end
+
       def default_fonts(options)
         { bodyfont: (if options[:script] == "Hans"
                        '"Source Han Sans",serif'
@@ -35,8 +44,7 @@ module IsoDoc
           header: html_doc_path("header.html"),
           wordcoverpage: html_doc_path("word_ieee_titlepage.html"),
           wordintropage: html_doc_path("word_ieee_intro.html"),
-          ulstyle: "l3",
-          olstyle: "l2" }
+          ulstyle: "l11", olstyle: "l16" }
       end
 
       def abstract(isoxml, out)
@@ -65,6 +73,94 @@ module IsoDoc
       def middle_title_ieee(_docxml, out)
         out.p(**{ class: "IEEEStdsTitle", style: "margin-top:70.0pt" }) do |p|
           p << @meta.get[:full_doctitle]
+          @meta.get[:amd] || @meta.get[:corr] and p << "<br/>"
+          @meta.get[:amd] and p << "Amendment #{@meta.get[:amd]}"
+          @meta.get[:amd] && @meta.get[:corr] and p << " "
+          @meta.get[:corr] and p << "Corrigenda #{@meta.get[:corr]}"
+        end
+      end
+
+      def admonition_name_parse(_node, div, name)
+        div.p **{ class: "IEEEStdsWarning", style: "text-align:center;" } do |p|
+          p.b do |b|
+            name.children.each { |n| parse(n, b) }
+          end
+        end
+      end
+
+      def admonition_class(node)
+        if node["type"] == "editorial" then "zzHelp"
+        elsif node.ancestors("introduction").empty?
+          "IEEEStdsWarning"
+        else "IEEEStdsIntroduction"
+        end
+      end
+
+      def dt_dd?(node)
+        %w{dt dd}.include? node.name
+      end
+
+      def formula_where(dlist, out)
+        return unless dlist
+
+        dlist.elements.select { |n| dt_dd? n }.each_slice(2) do |dt, dd|
+          formula_where1(out, dt, dd)
+        end
+      end
+
+      def formula_where1(out, dterm, ddefn)
+        out.p **{ class: "IEEEStdsEquationVariableList" } do |p|
+          dterm.children.each { |n| parse(n, p) }
+          insert_tab(p, 1)
+          if ddefn.at(ns("./p"))
+            ddefn.elements.each do |e|
+              e.children.each { |n| parse(n, p) }
+            end
+          else ddefn.children.each { |n| parse(n, p) }
+          end
+        end
+      end
+
+      def annex_attrs(node)
+        { id: node["id"], class: "Annex" }
+      end
+
+      def annex_name(_annex, name, div)
+        preceding_floating_titles(name, div)
+        return if name.nil?
+
+        name&.at(ns("./strong"))&.remove # supplied by CSS list numbering
+        div.h1 **{ class: "Annex" } do |t|
+          annex_name1(name, t)
+          clause_parse_subtitle(name, t)
+        end
+      end
+
+      def annex_name1(name, out)
+        name.children.each do |c2|
+          if c2.name == "span" && c2["class"] == "obligation"
+            out.span **{ style: "font-weight:normal;" } do |s|
+              c2.children.each { |c3| parse(c3, s) }
+            end
+          else parse(c2, out)
+          end
+        end
+      end
+
+      def termnote_parse(node, out)
+        name = node&.at(ns("./name"))&.remove
+        out.div **note_attrs(node) do |div|
+          div.p do |p|
+            name and termnote_label(p, name)
+            para_then_remainder(node.first_element_child, node, p, div)
+          end
+        end
+      end
+
+      def termnote_label(para, name)
+        para.span **{ class: "note_label" } do |s|
+          name.children.each { |n| parse(n, s) }
+          s << termnote_delim
         end
       end
 

data/lib/isodoc/ieee/xref.rb

@@ -72,6 +72,37 @@ module IsoDoc
           )
         end
       end
+
+      def termnote_anchor_names(docxml)
+        docxml.xpath(ns("//*[termnote]")).each do |t|
+          c = Counter.new
+          sequence = UUIDTools::UUID.random_create.to_s
+          notes = t.xpath(ns("./termnote"))
+          notes.reject { |n| blank?(n["id"]) }.each do |n|
+            @anchors[n["id"]] =
+              anchor_struct("#{@labels['termnote']} #{increment_label(notes, n, c)}",
+                            n,
+                            @labels["note_xref"], "termnote", false)
+                .merge(sequence: sequence)
+          end
+        end
+      end
+
+      def note_anchor_names1(notes, counter)
+        sequence = UUIDTools::UUID.random_create.to_s
+        notes.each do |n|
+          next if @anchors[n["id"]] || blank?(n["id"])
+
+          @anchors[n["id"]] =
+            anchor_struct(increment_label(notes, n, counter), n,
+                          @labels["note_xref"], "note", false)
+              .merge(sequence: sequence)
+        end
+      end
+
+      def annex_name_lbl(clause, num)
+        super.sub(%r{<br/>(.*)$}, "<br/><span class='obligation'>\\1</span>")
+      end
     end
   end
 end