isodoc 0.5.5 → 0.5.7

data/lib/isodoc/section.rb

@@ -3,7 +3,7 @@ module IsoDoc
     def inline_header_title(out, node, c1)
       out.span **{ class: "zzMoveToFollowing" } do |s|
         s.b do |b|
-          b << "#{get_anchors[node['id']][:label]}. #{c1.text} "
+          b << "#{get_anchors[node['id']][:label]}. #{c1.content} "
         end
       end
     end
@@ -13,7 +13,8 @@ module IsoDoc
         inline_header_title(out, node, c1)
       else
         div.send "h#{get_anchors[node['id']][:level]}" do |h|
-          h << "#{get_anchors[node['id']][:label]}. #{c1.text}"
+          h << "#{get_anchors[node['id']][:label]}. "
+          c1.children.each { |c2| parse(c2, h) }
         end
       end
     end
@@ -49,7 +50,7 @@ module IsoDoc
           c.elements.each do |c1|
             if c1.name == "title"
               clause_name("#{get_anchors[c['id']][:label]}.",
-                          c1.text, s, nil)
+                          c1.content, s, nil)
             else
               parse(c1, s)
             end

data/lib/isodoc/table.rb

@@ -76,9 +76,7 @@ module IsoDoc
       rowmax = td["rowspan"] ? row + td["rowspan"].to_i - 1 : row
       style += <<~STYLE
         border-top:#{row.zero? ? "#{SW} 1.5pt;" : 'none;'}
-        mso-border-top-alt:#{row.zero? ? "#{SW} 1.5pt;" : 'none;'}
         border-bottom:#{SW} #{rowmax == totalrows ? '1.5' : '1.0'}pt;
-        mso-border-bottom-alt:#{SW} #{rowmax == totalrows ? '1.5' : '1.0'}pt;
       STYLE
       { rowspan: td["rowspan"], colspan: td["colspan"],
         align: td["align"], style: style.gsub(/\n/, "") }

data/lib/isodoc/terms.rb

@@ -24,7 +24,7 @@ module IsoDoc
       out.p **{ class: "Terms" } { |p| p << node.text }
     end
 
-    def para_then_remainder(first, node, p)
+    def para_then_remainder(first, node, p, div)
       if first.name == "p"
         first.children.each { |n| parse(n, p) }
         node.elements.drop(1).each { |n| parse(n, div) }
@@ -33,23 +33,12 @@ module IsoDoc
       end
     end
 
-    def termexample_parse(node, out)
-      out.div **{ class: "Note" } do |div|
-        first = node.first_element_child
-        div.p **{ class: "Note" } do |p|
-          p << l10n("#{@example_lbl}:")
-          insert_tab(p, 1)
-          para_then_remainder(first, node, p)
-        end
-      end
-    end
-
     def termnote_parse(node, out)
       out.div **{ class: "Note" } do |div|
         first = node.first_element_child
         div.p **{ class: "Note" } do |p|
           p << "#{get_anchors[node['id']][:label]}: "
-          para_then_remainder(first, node, p)
+          para_then_remainder(first, node, p, div)
         end
       end
     end

data/lib/isodoc/utils.rb

@@ -8,9 +8,7 @@ module IsoDoc
     end
 
     def insert_tab(out, n)
-      out.span **attr_code(style: "mso-tab-count:#{n}") do |span|
-        [1..n].each { span << "&#xA0; " }
-      end
+      [1..n].each { out << "&nbsp; " }
     end
 
     STAGE_ABBRS = {
@@ -119,5 +117,28 @@ module IsoDoc
       end
       [@openmathdelim, @closemathdelim]
     end
+
+    def header_strip(h)
+      h = h.to_s.gsub(%r{<br/>}, " ").sub(/<\/?h[12][^>]*>/, "")
+      h1 = to_xhtml_fragment(h.dup)
+      h1.traverse do |x|
+        x.remove if x.name == "span" && x["class"] == "MsoCommentReference"
+        x.remove if x.name == "a" && x["epub:type"] == "footnote"
+        if x.name == "a"
+          x.replace(x.children)
+        end
+      end
+      from_xhtml(h1)
+    end
+
+    def populate_template(docxml, _format)
+      meta = get_metadata
+      docxml = docxml.
+        gsub(/\[TERMREF\]\s*/, l10n("[#{@source_lbl}: ")).
+        gsub(/\s*\[\/TERMREF\]\s*/, l10n("]")).
+        gsub(/\s*\[MODIFICATION\]/, l10n(", #{@modified_lbl} &mdash; "))
+      template = Liquid::Template.parse(docxml)
+      template.render(meta.map { |k, v| [k.to_s, v] }.to_h)
+    end
   end
 end

data/lib/isodoc/version.rb

@@ -1,3 +1,3 @@
 module IsoDoc
-  VERSION = "0.5.5".freeze
+  VERSION = "0.5.7".freeze
 end

data/lib/isodoc/wordconvert/comments.rb

@@ -0,0 +1,155 @@
+#require "uuidtools"
+
+#module IsoDoc
+  #class WordConvert
+  #module WordConvertModule
+    #def self.included base
+#base.class_eval do
+
+def in_comment
+  @in_comment
+end
+
+def comments(div)
+  return if @comments.empty?
+  div.div **{ style: "mso-element:comment-list" } do |div1|
+    @comments.each { |fn| div1.parent << fn }
+  end
+end
+
+def review_note_parse(node, out)
+  fn = @comments.length + 1
+  make_comment_link(out, fn, node)
+  @in_comment = true
+  @comments << make_comment_text(node, fn)
+  @in_comment = false
+end
+
+def comment_link_attrs(fn, node)
+  { style: "MsoCommentReference", target: fn,
+    class: "commentLink", from: node["from"],
+    to: node["to"] }
+end
+
+# add in from and to links to move the comment into place
+def make_comment_link(out, fn, node)
+  out.span(**comment_link_attrs(fn, node)) do |s1|
+    s1.span **{ lang: "EN-GB", style: "font-size:9.0pt" } do |s2|
+      s2.a **{ style: "mso-comment-reference:SMC_#{fn};"\
+               "mso-comment-date:#{node['date']}" }
+      s2.span **{ style: "mso-special-character:comment",
+                  target: fn } # 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" }
+    end
+  end
+end
+
+def make_comment_text(node, fn)
+  noko do |xml|
+    xml.div **{ style: "mso-element:comment", id: fn } do |div|
+      div.span **{ style: %{mso-comment-author:"#{node['reviewer']}"} }
+      make_comment_target(div)
+      node.children.each { |n| parse(n, div) }
+    end
+  end.join("\n")
+end
+
+def comment_cleanup(docxml)
+  move_comment_link_to_from(docxml)
+  reorder_comments_by_comment_link(docxml)
+  embed_comment_in_comment_list(docxml)
+end
+
+#COMMENT_IN_COMMENT_LIST =
+COMMENT_IN_COMMENT_LIST1 =
+  '//div[@style="mso-element:comment-list"]//'\
+  'span[@style="MsoCommentReference"]'.freeze
+
+def embed_comment_in_comment_list(docxml)
+  #docxml.xpath(COMMENT_IN_COMMENT_LIST).each do |x|
+  docxml.xpath(COMMENT_IN_COMMENT_LIST1).each do |x|
+    n = x.next_element
+    n&.children&.first&.add_previous_sibling(x.remove)
+  end
+  docxml
+end
+
+def move_comment_link_to_from1(x, fromlink)
+  x.remove
+  link = x.at(".//a")
+  fromlink.replace(x)
+  link.children = fromlink
+end
+
+def comment_attributes(docxml, x)
+  fromlink = docxml.at("//*[@id='#{x['from']}']")
+  return(nil) if fromlink.nil?
+  tolink = docxml.at("//*[@id='#{x['to']}']") || fromlink
+  target = docxml.at("//*[@id='#{x['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
+end
+
+def skip_comment_wrap(from)
+  from["style"] != "mso-special-character:comment"
+end
+
+def insert_comment_cont(from, to, target)
+  # includes_to = from.at(".//*[@id='#{to}']")
+  while !from.nil? && from["id"] != to
+    following = from.xpath("./following::*")
+    (from = following.shift) && incl_to = from.at(".//*[@id='#{to}']")
+    while !incl_to.nil? && !from.nil? && skip_comment_wrap(from)
+      (from = following.shift) && incl_to = from.at(".//*[@id='#{to}']")
+    end
+    wrap_comment_cont(from, target) if !from.nil?
+  end
+end
+
+def move_comment_link_to_from(docxml)
+  docxml.xpath('//span[@style="MsoCommentReference"][@from]').each do |x|
+    attrs = comment_attributes(docxml, x) || next
+    move_comment_link_to_from1(x, attrs[:from])
+    insert_comment_cont(attrs[:from], x["to"], x["target"])
+  end
+end
+
+def get_comments_from_text(docxml, link_order)
+  comments = []
+  docxml.xpath("//div[@style='mso-element:comment']").each do |c|
+    next unless c["id"] && !link_order[c["id"]].nil?
+    comments << { text: c.remove.to_s, id: c["id"] }
+  end
+  comments.sort! { |a, b| link_order[a[:id]] <=> link_order[b[:id]] }
+  # comments
+end
+
+#COMMENT_TARGET_XREFS =
+COMMENT_TARGET_XREFS1 =
+  "//span[@style='mso-special-character:comment']/@target".freeze
+
+def reorder_comments_by_comment_link(docxml)
+  link_order = {}
+  #docxml.xpath(COMMENT_TARGET_XREFS).each_with_index do |target, i|
+  docxml.xpath(COMMENT_TARGET_XREFS1).each_with_index do |target, i|
+    link_order[target.value] = i
+  end
+  comments = get_comments_from_text(docxml, link_order)
+  list = docxml.at("//*[@style='mso-element:comment-list']") || return
+  list.children = comments.map { |c| c[:text] }.join("\n")
+end
+      #end
+    #end
+  #end
+#end

data/lib/isodoc/wordconvert/convert.rb

@@ -0,0 +1,31 @@
+require "uuidtools"
+require "html2doc"
+require "liquid"
+
+require_relative "wordconvertmodule"
+require_relative "comments"
+require_relative "footnotes"
+
+module IsoDoc
+
+  module WordConvertModule
+    # http://tech.tulentsev.com/2012/02/ruby-how-to-override-class-method-with-a-module/
+    # https://www.ruby-forum.com/topic/148303
+    #
+    # The following is ugly indeed, but the only way I can split module override methods
+    # across files
+    def self.included base
+      base.class_eval do
+
+        eval File.open(File.join(File.dirname(__FILE__),"wordconvertmodule.rb")).read
+        eval File.open(File.join(File.dirname(__FILE__),"comments.rb")).read
+        eval File.open(File.join(File.dirname(__FILE__),"footnotes.rb")).read
+      end
+    end
+  end
+
+  class WordConvert < Convert
+    include WordConvertModule
+  end
+end
+

data/lib/isodoc/wordconvert/footnotes.rb

@@ -0,0 +1,80 @@
+#require "uuidtools"
+
+#module IsoDoc
+  #class WordConvert
+  #module WordConvertModule
+    #def self.included base
+#base.class_eval do
+
+def footnotes(div)
+  return if @footnotes.empty?
+  @footnotes.each { |fn| div.parent << fn }
+end
+
+def make_table_footnote_link(out, fnid, fnref)
+  attrs = { href: "##{fnid}", class: "TableFootnoteRef" }
+  out.a **attrs do |a|
+    a << fnref
+  end
+end
+
+def make_table_footnote_target(out, fnid, fnref)
+  attrs = { id: fnid, class: "TableFootnoteRef" }
+  out.a **attrs do |a|
+    a << fnref
+    insert_tab(a, 1)
+  end
+end
+
+def make_table_footnote_text(node, fnid, fnref)
+  attrs = { id: "ftn#{fnid}" }
+  noko do |xml|
+    xml.div **attr_code(attrs) do |div|
+      make_table_footnote_target(div, fnid, fnref)
+      node.children.each { |n| parse(n, div) }
+    end
+  end.join("\n")
+end
+
+def make_generic_footnote_text(node, fnid)
+  noko do |xml|
+    xml.aside **{ id: "ftn#{fnid}" } do |div|
+      node.children.each { |n| parse(n, div) }
+    end
+  end.join("\n")
+end
+
+def get_table_ancestor_id(node)
+  table = node.ancestors("table") || node.ancestors("figure")
+  return UUIDTools::UUID.random_create.to_s if table.empty?
+  table.last["id"]
+end
+
+def table_footnote_parse(node, out)
+  fn = node["reference"]
+  tid = get_table_ancestor_id(node)
+  make_table_footnote_link(out, tid + fn, fn)
+  # do not output footnote text if we have already seen it for this table
+  return if @seen_footnote.include?(tid + fn)
+  @in_footnote = true
+  out.aside { |a| a << make_table_footnote_text(node, tid + fn, fn) }
+  @in_footnote = false
+  @seen_footnote << (tid + fn)
+end
+
+def footnote_parse(node, out)
+  return table_footnote_parse(node, out) if @in_table || @in_figure
+  fn = node["reference"]
+  out.a **{ "epub:type": "footnote", href: "#ftn#{fn}" } do |a|
+    a.sup { |sup| sup << fn }
+  end
+  return if @seen_footnote.include?(fn)
+  @in_footnote = true
+  @footnotes << make_generic_footnote_text(node, fn)
+  @in_footnote = false
+  @seen_footnote << fn
+end
+      #end
+    #end
+  #end
+#end

data/lib/isodoc/wordconvert/wordconvertmodule.rb

@@ -0,0 +1,212 @@
+
+#require "html2doc"
+#require "liquid"
+
+#module IsoDoc
+#class WordConvert < Convert
+#module WordConvertModule
+#def self.included base 
+#base.class_eval do
+def insert_tab(out, n)
+  out.span **attr_code(style: "mso-tab-count:#{n}") do |span|
+    [1..n].each { span << "&#xA0; " }
+  end
+end
+
+def para_attrs(node)
+  classtype = nil
+  classtype = "Note" if @note
+  classtype = "MsoCommentText" if in_comment
+  classtype = "Sourcecode" if @annotation
+  attrs = { class: classtype, id: node["id"] }
+  unless node["align"].nil?
+    attrs[:align] = node["align"] unless node["align"] == "justify"
+    attrs[:style] = "text-align:#{node['align']}"
+  end
+  attrs
+end
+
+def remove_bottom_border(td)
+  td["style"] =
+    td["style"].gsub(/border-bottom:[^;]+;/, "border-bottom:0pt;").
+    gsub(/mso-border-bottom-alt:[^;]+;/, "mso-border-bottom-alt:0pt;")
+end
+
+#SW1 = IsoDoc::SW
+SW1 = "solid windowtext".freeze
+
+def new_fullcolspan_row(t, tfoot)
+  # how many columns in the table?
+  cols = 0
+  t.at(".//tr").xpath("./td | ./th").each do |td|
+    cols += (td["colspan"] ? td["colspan"].to_i : 1)
+  end
+  style = %{border-top:0pt;mso-border-top-alt:0pt;
+      border-bottom:#{SW1} 1.5pt;mso-border-bottom-alt:#{SW1} 1.5pt;}
+  tfoot.add_child("<tr><td colspan='#{cols}' style='#{style}'/></tr>")
+  tfoot.xpath(".//td").last
+end
+
+def make_tr_attr(td, row, totalrows)
+  style = td.name == "th" ? "font-weight:bold;" : ""
+  rowmax = td["rowspan"] ? row + td["rowspan"].to_i - 1 : row
+  style += <<~STYLE
+        border-top:#{row.zero? ? "#{SW1} 1.5pt;" : 'none;'}
+        mso-border-top-alt:#{row.zero? ? "#{SW1} 1.5pt;" : 'none;'}
+        border-bottom:#{SW1} #{rowmax == totalrows ? '1.5' : '1.0'}pt;
+        mso-border-bottom-alt:#{SW1} #{rowmax == totalrows ? '1.5' : '1.0'}pt;
+  STYLE
+  { rowspan: td["rowspan"], colspan: td["colspan"],
+    align: td["align"], style: style.gsub(/\n/, "") }
+end
+
+def section_break(body)
+  body.br **{ clear: "all", class: "section" }
+end
+
+def page_break(out)
+  out.br **{
+    clear: "all",
+    style: "mso-special-character:line-break;page-break-before:always",
+  }
+end
+
+def dt_parse(dt, term)
+  if dt.elements.empty?
+    term.p **attr_code(class: note? ? "Note" : nil,
+                       style: "text-align: left;") do |p|
+      p << dt.text
+    end
+  else
+    dt.children.each { |n| parse(n, term) }
+  end
+end
+
+def dl_parse(node, out)
+  out.table **{ class: "dl" } do |v|
+    node.elements.each_slice(2) do |dt, dd|
+      v.tr do |tr|
+        tr.td **{ valign: "top", align: "left" } do |term|
+          dt_parse(dt, term)
+        end
+        tr.td **{ valign: "top" } do |listitem|
+          dd.children.each { |n| parse(n, listitem) }
+        end
+      end
+    end
+  end
+end
+
+
+
+def postprocess(result, filename, dir)
+  generate_header(filename, dir)
+  result = from_xhtml(cleanup(to_xhtml(result)))
+  toWord(result, filename, dir)
+end
+
+def toWord(result, filename, dir)
+  result = from_xhtml(word_cleanup(to_xhtml(result)))
+  result = populate_template(result, :word)
+  Html2Doc.process(result, filename: filename, stylesheet: @wordstylesheet,
+                   header_file: "header.html", dir: dir,
+                   asciimathdelims: [@openmathdelim, @closemathdelim],
+                   liststyles: { ul: @ulstyle, ol: @olstyle })
+end
+
+def word_cleanup(docxml)
+  word_preface(docxml)
+  word_annex_cleanup(docxml)
+  docxml
+end
+
+# force Annex h2 to be p.h2Annex, so it is not picked up by ToC
+def word_annex_cleanup(docxml)
+  docxml.xpath("//h2[ancestor::*[@class = 'Section3']]").each do |h2|
+    h2.name = "p"
+    h2["class"] = "h2Annex"
+  end
+end
+
+def word_preface(docxml)
+  word_cover(docxml) if @wordcoverpage
+  word_intro(docxml) if @wordintropage
+end
+
+def word_cover(docxml)
+  cover = to_xhtml_fragment(File.read(@wordcoverpage, encoding: "UTF-8"))
+  docxml.at('//div[@class="WordSection1"]').children.first.previous =
+    cover.to_xml(encoding: "US-ASCII")
+end
+
+def word_intro(docxml)
+  intro = to_xhtml_fragment(File.read(@wordintropage, encoding: "UTF-8").
+                            sub(/WORDTOC/, make_WordToC(docxml)))
+  docxml.at('//div[@class="WordSection2"]').children.first.previous =
+    intro.to_xml(encoding: "US-ASCII")
+end
+
+def generate_header(filename, _dir)
+  return unless @header
+  template = Liquid::Template.parse(File.read(@header, encoding: "UTF-8"))
+  meta = get_metadata
+  meta[:filename] = filename
+  params = meta.map { |k, v| [k.to_s, v] }.to_h
+  File.open("header.html", "w") do |f|
+    f.write(template.render(params))
+  end
+end
+
+def word_toc_entry(toclevel, heading)
+  bookmark = 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
+
+#WORD_TOC_PREFACE = <<~TOC.freeze
+WORD_TOC_PREFACE1 = <<~TOC.freeze
+      <span lang="EN-GB"><span
+        style='mso-element:field-begin'></span><span
+        style='mso-spacerun:yes'>&#xA0;</span>TOC
+        \\o &quot;1-2&quot; \\h \\z \\u <span
+        style='mso-element:field-separator'></span></span>
+TOC
+
+#WORD_TOC_SUFFIX = <<~TOC.freeze
+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>&nbsp;</o:p></span></p>
+TOC
+
+def make_WordToC(docxml)
+  toc = ""
+  docxml.xpath("//h1 | //h2[not(ancestor::*[@class = 'Section3'])]").
+    each do |h|
+    toc += word_toc_entry(h.name == "h1" ? 1 : 2, header_strip(h))
+  end
+  toc.sub(/(<p class="MsoToc1">)/,
+          #%{\\1#{WORD_TOC_PREFACE}}) + WORD_TOC_SUFFIX
+          %{\\1#{WORD_TOC_PREFACE1}}) +  WORD_TOC_SUFFIX1
+end
+#end
+#end
+#end
+#end