isodoc 2.4.1 → 2.4.3

data/lib/isodoc/word_function/comments.rb

@@ -4,7 +4,7 @@ module IsoDoc
       def comments(div)
         return if @comments.empty?
 
-        div.div **{ style: "mso-element:comment-list" } do |div1|
+        div.div style: "mso-element:comment-list" do |div1|
           @comments.each { |fn| div1.parent << fn }
         end
       end
@@ -26,28 +26,27 @@ module IsoDoc
       # add in from and to links to move the comment into place
       def make_comment_link(out, fnote, node)
         out.span(**comment_link_attrs(fnote, node)) do |s1|
-          s1.span **{ lang: "EN-GB", style: "font-size:9.0pt" } do |s2|
-            s2.a **{ style: "mso-comment-reference:SMC_#{fnote};"\
+          s1.span lang: "EN-GB", style: "font-size:9.0pt" do |s2|
+            s2.a style: "mso-comment-reference:SMC_#{fnote};" \
                             "mso-comment-date:#{node['date'].gsub(/[:-]+/,
-                                                                  '')}" }
-            s2.span **{ style: "mso-special-character:comment",
-                        target: fnote } # do |s|
+                                                                  '')}"
+            s2.span style: "mso-special-character:comment", target: fnote # do |s|
           end
         end
       end
 
       def make_comment_target(out)
-        out.span **{ style: "MsoCommentReference" } do |s1|
-          s1.span **{ lang: "EN-GB", style: "font-size:9.0pt" } do |s2|
-            s2.span **{ style: "mso-special-character:comment" }
+        out.span style: "MsoCommentReference" do |s1|
+          s1.span lang: "EN-GB", style: "font-size:9.0pt" do |s2|
+            s2.span style: "mso-special-character:comment"
           end
         end
       end
 
       def make_comment_text(node, fnote)
         noko do |xml|
-          xml.div **{ style: "mso-element:comment", id: fnote } do |div|
-            div.span **{ style: %{mso-comment-author:"#{node['reviewer']}"} }
+          xml.div style: "mso-element:comment", id: fnote do |div|
+            div.span style: %{mso-comment-author:"#{node['reviewer']}"}
             make_comment_target(div)
             node.children.each { |n| parse(n, div) }
           end
@@ -61,7 +60,7 @@ module IsoDoc
       end
 
       COMMENT_IN_COMMENT_LIST1 =
-        '//div[@style="mso-element:comment-list"]//'\
+        '//div[@style="mso-element:comment-list"]//' \
         'span[@style="MsoCommentReference"]'.freeze
 
       def embed_comment_in_comment_list(docxml)
@@ -80,18 +79,24 @@ module IsoDoc
         link.children = fromlink
       end
 
-      def comment_attributes(docxml, x)
-        fromlink = docxml.at("//*[@id='#{x['from']}']")
+      def comment_attributes(docxml, span)
+        fromlink = docxml.at("//*[@id='#{span['from']}']")
         return(nil) if fromlink.nil?
 
-        tolink = docxml.at("//*[@id='#{x['to']}']") || fromlink
-        target = docxml.at("//*[@id='#{x['target']}']")
+        tolink = docxml.at("//*[@id='#{span['to']}']") || fromlink
+        target = docxml.at("//*[@id='#{span['target']}']")
         { from: fromlink, to: tolink, target: target }
       end
 
       def wrap_comment_cont(from, target)
-        s = from.replace("<span style='mso-comment-continuation:#{target}'>")
-        s.first.children = from
+        if %w(ol ul li div p).include?(from.name)
+          from.children.each do |c|
+            wrap_comment_cont(c, target)
+          end
+        else
+          s = from.replace("<span style='mso-comment-continuation:#{target}'>")
+          s.first.children = from
+        end
       end
 
       def skip_comment_wrap(from)

data/lib/isodoc/word_function/footnotes.rb

@@ -45,7 +45,7 @@ module IsoDoc
 
       def make_generic_footnote_text(node, fnid)
         noko do |xml|
-          xml.aside **{ id: "ftn#{fnid}" } do |div|
+          xml.aside id: "ftn#{fnid}" do |div|
             node.children.each { |n| parse(n, div) }
           end
         end.join("\n")
@@ -72,13 +72,13 @@ module IsoDoc
       end
 
       def seen_footnote_parse(_node, out, footnote)
-        out.span **{ style: "mso-element:field-begin" }
+        out.span style: "mso-element:field-begin"
         out << " NOTEREF _Ref#{@fn_bookmarks[footnote]} \\f \\h"
-        out.span **{ style: "mso-element:field-separator" }
-        out.span **{ class: "MsoFootnoteReference" } do |s|
+        out.span style: "mso-element:field-separator"
+        out.span class: "MsoFootnoteReference" do |s|
           s << footnote
         end
-        out.span **{ style: "mso-element:field-end" }
+        out.span style: "mso-element:field-end"
       end
 
       def footnote_parse(node, out)
@@ -89,9 +89,9 @@ module IsoDoc
         return seen_footnote_parse(node, out, fn) if @seen_footnote.include?(fn)
 
         @fn_bookmarks[fn] = bookmarkid
-        out.span **{ style: "mso-bookmark:_Ref#{@fn_bookmarks[fn]}" } do |s|
-          s.a **{ class: "FootnoteRef", "epub:type": "footnote",
-                  href: "#ftn#{fn}" } do |a|
+        out.span style: "mso-bookmark:_Ref#{@fn_bookmarks[fn]}" do |s|
+          s.a class: "FootnoteRef", "epub:type": "footnote",
+              href: "#ftn#{fn}" do |a|
             a.sup { |sup| sup << fn }
           end
         end

data/lib/isodoc/word_function/postprocess.rb

@@ -1,34 +1,11 @@
 require "fileutils"
 require_relative "./postprocess_cover"
+require_relative "./postprocess_toc"
+require_relative "./postprocess_table"
 
 module IsoDoc
   module WordFunction
     module Postprocess
-      # add namespaces for Word fragments
-      WORD_NOKOHEAD = <<~HERE.freeze
-        <!DOCTYPE html SYSTEM "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
-        <html xmlns="http://www.w3.org/1999/xhtml"
-        xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office"
-        xmlns:w="urn:schemas-microsoft-com:office:word"
-        xmlns:m="http://schemas.microsoft.com/office/2004/12/omml">
-        <head> <title></title> <meta charset="UTF-8" /> </head>
-        <body> </body> </html>
-      HERE
-
-      def to_word_xhtml_fragment(xml)
-        doc = ::Nokogiri::XML.parse(WORD_NOKOHEAD)
-        ::Nokogiri::XML::DocumentFragment.new(doc, xml, doc.root)
-      end
-
-      def table_note_cleanup(docxml)
-        super
-        # preempt html2doc putting MsoNormal there
-        docxml.xpath("//p[not(self::*[@class])][ancestor::*[@class = 'Note']]")
-          .each do |p|
-          p["class"] = "Note"
-        end
-      end
-
       def postprocess(result, filename, dir)
         filename = filename.sub(/\.doc$/, "")
         header = generate_header(filename, dir)
@@ -56,23 +33,6 @@ module IsoDoc
         "Sourcecode"
       end
 
-      def wordstylesheet_update
-        return if @wordstylesheet.nil?
-
-        f = File.open(@wordstylesheet.path, "a")
-        @landscapestyle.empty? or f.write(@landscapestyle)
-        s = @meta.get[:code_css] and
-          f.write(s.gsub(/sourcecode/, "p.#{sourcecode_style}"))
-        if @wordstylesheet_override && @wordstylesheet
-          f.write(@wordstylesheet_override.read)
-          @wordstylesheet_override.close
-        elsif @wordstylesheet_override && !@wordstylesheet
-          @wordstylesheet = @wordstylesheet_override
-        end
-        f.close
-        @wordstylesheet
-      end
-
       def word_admonition_images(docxml)
         docxml.xpath("//div[@class = 'Admonition']//img").each do |i|
           i["width"], i["height"] =
@@ -84,10 +44,12 @@ module IsoDoc
       def word_cleanup(docxml)
         word_annex_cleanup(docxml)
         word_preface(docxml)
+        word_sourcecode_annotations(docxml)
         word_sourcecode_table(docxml)
         word_nested_tables(docxml)
         word_colgroup(docxml)
         word_table_align(docxml)
+        word_table_pagebreak(docxml)
         word_table_separator(docxml)
         word_admonition_images(docxml)
         word_list_continuations(docxml)
@@ -101,18 +63,29 @@ module IsoDoc
         docxml
       end
 
-      def word_sourcecode_table(docxml)
-        docxml.xpath("//p[@class='Sourcecode']/div[@class='table_container']")
-          .each do |d|
-            pre = d.at(".//p[@class='Sourcecode']")
-            to_sourcecode_para(pre)
-            d["id"] = d.parent["id"]
-            d.parent.replace(d)
+      def word_sourcecode_annotations(html)
+        ann = ".//div[@class = 'annotation']"
+        html.xpath("//p[@class = '#{sourcecode_style}'][#{ann}]")
+          .each do |p|
+          ins = p.after("<p class='#{sourcecode_style}'/>").next_element
+          p.xpath(ann).each do |d|
+            ins << d.remove.children
           end
+        end
+      end
+
+      def word_sourcecode_table(docxml)
+        s = "p[@class='#{sourcecode_style}']"
+        docxml.xpath("//#{s}/div[@class='table_container']").each do |d|
+          pre = d.at(".//#{s}")
+          to_sourcecode_para(pre)
+          d["id"] = d.parent["id"]
+          d.parent.replace(d)
+        end
       end
 
       def to_sourcecode_para(pre)
-        @sourcecode = true
+        @sourcecode = "pre"
         pre.traverse do |x|
           x.text? or next
           ret = []
@@ -130,63 +103,14 @@ module IsoDoc
         end
       end
 
-      def word_colgroup(docxml)
-        cells2d = {}
-        docxml.xpath("//table[colgroup]").each do |t|
-          w = colgroup_widths(t)
-          t.xpath(".//tr").each_with_index { |_tr, r| cells2d[r] = {} }
-          t.xpath(".//tr").each_with_index do |tr, r|
-            tr.xpath("./td | ./th").each_with_index do |td, _i|
-              x = 0
-              rs = td.attr("rowspan")&.to_i || 1
-              cs = td.attr("colspan")&.to_i || 1
-              while cells2d[r][x]
-                x += 1
-              end
-              (r..(r + rs - 1)).each do |y2|
-                cells2d[y2].nil? and next
-                (x..(x + cs - 1)).each { |x2| cells2d[y2][x2] = 1 }
-              end
-              width = (x..(x + cs - 1)).each_with_object({ width: 0 }) do |z, m|
-                m[:width] += w[z]
-              end
-              td["width"] = "#{width[:width]}%"
-              x += cs
-            end
-          end
-        end
-      end
-
-      # assume percentages
-      def colgroup_widths(table)
-        table.xpath("./colgroup/col").each_with_object([]) do |c, m|
-          m << c["width"].sub(/%$/, "").to_f
-        end
-      end
-
-      def word_nested_tables(docxml)
-        docxml.xpath("//table").each do |t|
-          t.xpath(".//table").reverse.each do |tt|
-            t.next = tt.remove
-          end
-        end
-      end
-
-      def style_update(node, css)
-        return unless node
-
-        node["style"] =
-          node["style"] ? node["style"].sub(/;?$/, ";#{css}") : css
-      end
-
       def word_image_caption(docxml)
         docxml.xpath("//p[@class = 'FigureTitle' or @class = 'SourceTitle']")
           .each do |t|
-          if t&.previous_element&.name == "img"
+          if t.previous_element&.name == "img"
             img = t.previous_element
             t.previous_element.swap("<p class='figure'>#{img.to_xml}</p>")
           end
-          style_update(t&.previous_element, "page-break-after:avoid;")
+          style_update(t.previous_element, "page-break-after:avoid;")
         end
       end
 
@@ -211,23 +135,6 @@ module IsoDoc
         end
       end
 
-      def word_table_align(docxml)
-        docxml.xpath("//td[@align]/p | //th[@align]/p").each do |p|
-          next if p["align"]
-
-          style_update(p, "text-align: #{p.parent['align']}")
-        end
-      end
-
-      def word_table_separator(docxml)
-        docxml.xpath("//p[@class = 'TableTitle']").each do |t|
-          next unless t.children.empty?
-
-          t["style"] = t["style"].sub(/;?$/, ";font-size:0pt;")
-          t.children = "&#xa0;"
-        end
-      end
-
       def word_annex_cleanup(docxml); end
 
       def word_example_cleanup(docxml)

data/lib/isodoc/word_function/postprocess_cover.rb

@@ -23,164 +23,36 @@ module IsoDoc
           introxml.to_xml(encoding: "US-ASCII")
       end
 
-      def insert_toc(intro, docxml, level)
-        toc = ""
-        toc += make_WordToC(docxml, level)
-        toc += make_table_word_toc(docxml)
-        toc += make_figure_word_toc(docxml)
-        toc += make_recommendation_word_toc(docxml)
-        intro.sub(/WORDTOC/, toc)
-      end
-
-      def word_toc_entry(toclevel, heading)
-        bookmark = bookmarkid # Random.rand(1000000000)
-        <<~TOC
-          <p class="MsoToc#{toclevel}"><span class="MsoHyperlink"><span lang="EN-GB" style='mso-no-proof:yes'>
-          <a href="#_Toc#{bookmark}">#{heading}<span lang="EN-GB" class="MsoTocTextSpan">
-          <span style='mso-tab-count:1 dotted'>. </span>
-          </span><span lang="EN-GB" class="MsoTocTextSpan">
-          <span style='mso-element:field-begin'></span></span>
-          <span lang="EN-GB" class="MsoTocTextSpan"> PAGEREF _Toc#{bookmark} \\h </span>
-            <span lang="EN-GB" class="MsoTocTextSpan"><span style='mso-element:field-separator'></span></span><span
-            lang="EN-GB" class="MsoTocTextSpan">1</span>
-            <span lang="EN-GB" class="MsoTocTextSpan"></span><span
-            lang="EN-GB" class="MsoTocTextSpan"><span style='mso-element:field-end'></span></span></a></span></span></p>
-
-        TOC
-      end
-
-      def word_toc_preface(level)
-        <<~TOC
-          <span lang="EN-GB"><span style='mso-element:field-begin'></span><span
-            style='mso-spacerun:yes'>&#xA0;</span>TOC \\o "1-#{level}" \\h \\z \\u <span
-            style='mso-element:field-separator'></span></span>
-        TOC
-      end
-
-      WORD_TOC_SUFFIX1 = <<~TOC.freeze
-        <p class="MsoToc1"><span lang="EN-GB"><span
-          style='mso-element:field-end'></span></span><span
-          lang="EN-GB"><o:p>&#xA0;</o:p></span></p>
-      TOC
-
-      def make_WordToC(docxml, level)
-        toc = ""
-        # docxml.xpath("//h1 | //h2[not(ancestor::*[@class = 'Section3'])]").
-        xpath = (1..level).each.map { |i| "//h#{i}" }.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
-
-      # inheriting gems need to add native Word name of style, if different
-      # including both CSS style name and human readable style name.
-      # Any human readable style name needs to come first for the Word template
-      # to work in regenerating the ToC
-      def table_toc_class
-        %w(TableTitle tabletitle)
-      end
-
-      def figure_toc_class
-        %w(FigureTitle figuretitle)
-      end
-
-      def reqt_toc_class
-        %w(RecommendationTitle RecommendationTestTitle
-           recommendationtitle recommendationtesttitle)
-      end
-
-      def toc_word_class_list(classes)
-        classes.map do |x|
-          / /.match?(x) ? %(&quot;#{x}&quot;) : x
-        end.join(",")
-      end
-
-      def word_toc_reqt_preface1
-        <<~TOC
-          <span lang="EN-GB"><span style='mso-element:field-begin'></span><span
-          style='mso-spacerun:yes'>&#xA0;</span>TOC \\h \\z \\t #{toc_word_class_list reqt_toc_class}
-          <span style='mso-element:field-separator'></span></span>
-        TOC
-      end
-
-      def word_toc_table_preface1
-        <<~TOC
-          <span lang="EN-GB"><span style='mso-element:field-begin'></span><span style='mso-spacerun:yes'>&#xA0;</span>TOC
-          \\h \\z \\t #{toc_word_class_list table_toc_class} <span style='mso-element:field-separator'></span></span>
-        TOC
-      end
-
-      def word_toc_figure_preface1
-        <<~TOC
-          <span lang="EN-GB"><span style='mso-element:field-begin'></span><span style='mso-spacerun:yes'>&#xA0;</span>TOC
-          \\h \\z \\t #{toc_word_class_list figure_toc_class} <span style='mso-element:field-separator'></span></span>
-        TOC
-      end
-
-      def table_toc_xpath
-        attr = table_toc_class.map { |x| "@class = '#{x}'" }
-        "//p[#{attr.join(' or ')}]"
-      end
-
-      def make_table_word_toc(docxml)
-        (docxml.at(table_toc_xpath) && @toctablestitle) or return ""
-        toc = %{<p class="TOCTitle">#{@toctablestitle}</p>}
-        docxml.xpath(table_toc_xpath).each do |h|
-          toc += word_toc_entry(1, header_strip(h))
-        end
-        toc.sub(/(<p class="MsoToc1">)/,
-                %{\\1#{word_toc_table_preface1}}) + WORD_TOC_SUFFIX1
-      end
-
-      def figure_toc_xpath
-        attr = figure_toc_class.map { |x| "@class = '#{x}'" }
-        "//p[#{attr.join(' or ')}]"
-      end
-
-      def make_figure_word_toc(docxml)
-        (docxml.at(figure_toc_xpath) && @tocfigurestitle) or return ""
-        toc = %{<p class="TOCTitle">#{@tocfigurestitle}</p>}
-        docxml.xpath(figure_toc_xpath).each do |h|
-          toc += word_toc_entry(1, header_strip(h))
-        end
-        toc.sub(/(<p class="MsoToc1">)/,
-                %{\\1#{word_toc_figure_preface1}}) + WORD_TOC_SUFFIX1
-      end
-
-      def reqt_toc_xpath
-        attr = reqt_toc_class.map { |x| "@class = '#{x}'" }
-        "//p[#{attr.join(' or ')}]"
-      end
-
-      def make_recommendation_word_toc(docxml)
-        (docxml.at(reqt_toc_xpath) && @tocrecommendationstitle) or return ""
-        toc = %{<p class="TOCTitle">#{@tocrecommendationstitle}</p>}
-        docxml.xpath(reqt_toc_xpath).sort_by do |h|
-          recommmendation_sort_key(h.text)
-        end.each do |h|
-          toc += word_toc_entry(1, header_strip(h))
-        end
-        toc.sub(/(<p class="MsoToc1">)/,
-                %{\\1#{word_toc_reqt_preface1}}) + WORD_TOC_SUFFIX1
-      end
-
-      def recommmendation_sort_key(header)
-        m = /^([^0-9]+) (\d+)/.match(header) || /^([^:]+)/.match(header)
-        m ||= [header, nil]
-        ret = "#{recommmendation_sort_key1(m[1])}::"
-        m[2] and ret += ("%04d" % m[2].to_i).to_s
-        ret
-      end
-
-      def recommmendation_sort_key1(type)
-        case type&.downcase
-        when "requirement" then "04"
-        when "recommendation" then "05"
-        when "permission" then "06"
-        else type
+      # add namespaces for Word fragments
+      WORD_NOKOHEAD = <<~HERE.freeze
+        <!DOCTYPE html SYSTEM "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
+        <html xmlns="http://www.w3.org/1999/xhtml"
+        xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office"
+        xmlns:w="urn:schemas-microsoft-com:office:word"
+        xmlns:m="http://schemas.microsoft.com/office/2004/12/omml">
+        <head> <title></title> <meta charset="UTF-8" /> </head>
+        <body> </body> </html>
+      HERE
+
+      def to_word_xhtml_fragment(xml)
+        doc = ::Nokogiri::XML.parse(WORD_NOKOHEAD)
+        ::Nokogiri::XML::DocumentFragment.new(doc, xml, doc.root)
+      end
+
+      def wordstylesheet_update
+        @wordstylesheet.nil? and return
+        f = File.open(@wordstylesheet.path, "a")
+        @landscapestyle.empty? or f.write(@landscapestyle)
+        s = @meta.get[:code_css] and
+          f.write(s.gsub(/sourcecode/, "p.#{sourcecode_style}"))
+        if @wordstylesheet_override && @wordstylesheet
+          f.write(@wordstylesheet_override.read)
+          @wordstylesheet_override.close
+        elsif @wordstylesheet_override && !@wordstylesheet
+          @wordstylesheet = @wordstylesheet_override
         end
+        f.close
+        @wordstylesheet
       end
 
       def authority_cleanup1(docxml, klass)

data/lib/isodoc/word_function/postprocess_table.rb

@@ -0,0 +1,85 @@
+module IsoDoc
+  module WordFunction
+    module Postprocess
+      def table_note_cleanup(docxml)
+        super
+        # preempt html2doc putting MsoNormal there
+        docxml.xpath("//p[not(self::*[@class])][ancestor::*[@class = 'Note']]")
+          .each { |p| p["class"] = "Note" }
+      end
+
+      def word_colgroup(docxml)
+        cells2d = {}
+        docxml.xpath("//table[colgroup]").each do |t|
+          w = colgroup_widths(t)
+          t.xpath(".//tr").each_with_index { |_tr, r| cells2d[r] = {} }
+          t.xpath(".//tr").each_with_index do |tr, r|
+            tr.xpath("./td | ./th").each_with_index do |td, _i|
+              x = 0
+              rs = td.attr("rowspan")&.to_i || 1
+              cs = td.attr("colspan")&.to_i || 1
+              while cells2d[r][x]
+                x += 1
+              end
+              (r..(r + rs - 1)).each do |y2|
+                cells2d[y2].nil? and next
+                (x..(x + cs - 1)).each { |x2| cells2d[y2][x2] = 1 }
+              end
+              width = (x..(x + cs - 1)).each_with_object({ width: 0 }) do |z, m|
+                m[:width] += w[z]
+              end
+              td["width"] = "#{width[:width]}%"
+              x += cs
+            end
+          end
+        end
+      end
+
+      # assume percentages
+      def colgroup_widths(table)
+        table.xpath("./colgroup/col").each_with_object([]) do |c, m|
+          m << c["width"].sub(/%$/, "").to_f
+        end
+      end
+
+      def word_nested_tables(docxml)
+        docxml.xpath("//table").each do |t|
+          t.xpath(".//table").reverse.each do |tt|
+            t.next = tt.remove
+          end
+        end
+      end
+
+      def style_update(node, css)
+        node or return
+        node["style"] =
+          node["style"] ? node["style"].sub(/;?$/, ";#{css}") : css
+      end
+
+      def word_table_align(docxml)
+        docxml.xpath("//td[@align]/p | //th[@align]/p").each do |p|
+          p["align"] and next
+          style_update(p, "text-align: #{p.parent['align']}")
+        end
+      end
+
+      def word_table_separator(docxml)
+        docxml.xpath("//p[@class = 'TableTitle']").each do |t|
+          t.children.empty? or next
+          t["style"] = t["style"].sub(/;?$/, ";font-size:0pt;")
+          t.children = "&#xa0;"
+        end
+      end
+
+      def word_table_pagebreak(docxml)
+        docxml.xpath("//td[@style] | //th[@style]").each do |t|
+          s = /(page-break-after:[^;]+)/.match(t["style"])
+          (s && s[1]) or next
+          t.xpath(".//div | .//p | .//pre").each do |p|
+            style_update(p, s[1])
+          end
+        end
+      end
+    end
+  end
+end