isodoc 1.2.6 → 1.4.0

data/lib/isodoc/presentation_function/inline.rb

@@ -1,3 +1,5 @@
+require "twitter_cldr"
+
 module IsoDoc
   class PresentationXMLConvert < ::IsoDoc::Convert
     def prefix_container(container, linkend, _target)
@@ -8,8 +10,8 @@ module IsoDoc
       if node["citeas"].nil? && node["bibitemid"]
         return @xrefs.anchor(node["bibitemid"] ,:xref) || "???"
       elsif node["target"] && node["droploc"]
-        return @xrefs.anchor(node["target"], :value) || 
-          @xrefs.anchor(node["target"], :label) || 
+        return @xrefs.anchor(node["target"], :value) ||
+          @xrefs.anchor(node["target"], :label) ||
           @xrefs.anchor(node["target"], :xref) || "???"
       elsif node["target"] && !/.#./.match(node["target"])
         linkend = anchor_linkend1(node)
@@ -38,7 +40,7 @@ module IsoDoc
     end
 
     def nearest_block_parent(node)
-      until %w(p title td th name formula 
+      until %w(p title td th name formula
         li dt dd sourcecode pre).include?(node.name)
         node = node.parent
       end
@@ -51,13 +53,13 @@ module IsoDoc
       end
     end
 
-    def get_linkend(node)
-      contents = non_locality_elems(node).select { |c| !c.text? || /\S/.match(c) }
+    def get_linkend(n)
+      contents = non_locality_elems(n).select { |c| !c.text? || /\S/.match(c) }
       return unless contents.empty?
-      link = anchor_linkend(node, docid_l10n(node["target"] || node["citeas"]))
-      link += eref_localities(node.xpath(ns("./locality | ./localityStack")), link)
-      non_locality_elems(node).each { |n| n.remove }
-      node.add_child(link)
+      link = anchor_linkend(n, docid_l10n(n["target"] || n["citeas"]))
+      link += eref_localities(n.xpath(ns("./locality | ./localityStack")), link)
+      non_locality_elems(n).each { |n| n.remove }
+      n.add_child(link)
     end
     # so not <origin bibitemid="ISO7301" citeas="ISO 7301">
     # <locality type="section"><reference>3.1</reference></locality></origin>
@@ -104,7 +106,8 @@ module IsoDoc
     # TODO: move to localization file
     def eref_localities1(target, type, from, to, delim, lang = "en")
       return "" if type == "anchor"
-      return l10n(eref_localities1_zh(target, type, from, to, delim)) if lang == "zh"
+      lang == "zh" and
+        return l10n(eref_localities1_zh(target, type, from, to, delim))
       ret = delim
       loc = @i18n.locality[type] || type.sub(/^locality:/, "").capitalize
       ret += " #{loc}"
@@ -114,31 +117,132 @@ module IsoDoc
     end
 
     def xref(docxml)
-      docxml.xpath(ns("//xref")).each do |f|
-        xref1(f)
-      end
+      docxml.xpath(ns("//xref")).each { |f| xref1(f) }
     end
 
     def eref(docxml)
-      docxml.xpath(ns("//eref")).each do |f|
-        xref1(f)
-      end
+      docxml.xpath(ns("//eref")).each { |f| xref1(f) }
     end
 
     def origin(docxml)
-      docxml.xpath(ns("//origin[not(termref)]")).each do |f|
-        xref1(f)
-      end
+      docxml.xpath(ns("//origin[not(termref)]")).each { |f| xref1(f) }
     end
 
     def quotesource(docxml)
-      docxml.xpath(ns("//quote/source")).each do |f|
-        xref1(f)
-      end
+      docxml.xpath(ns("//quote/source")).each { |f| xref1(f) }
     end
 
     def xref1(f)
       get_linkend(f)
     end
+
+    def concept(docxml)
+      docxml.xpath(ns("//concept")).each { |f| concept1(f) }
+    end
+
+    def concept1(node)
+      content = node.first_element_child.children.select do |c|
+        !%w{locality localityStack}.include? c.name
+      end.select { |c| !c.text? || /\S/.match(c) }
+      node.replace content.empty? ? 
+        @i18n.term_defined_in.sub(/%/, node.first_element_child.to_xml) :
+        "<em>#{node.children.to_xml}</em>"
+    end
+
+
+    MATHML = { "m" => "http://www.w3.org/1998/Math/MathML" }.freeze
+
+    def mathml(docxml)
+      locale = twitter_cldr_localiser()
+      docxml.xpath("//m:math", MATHML).each do |f|
+        mathml1(f, locale)
+      end
+    end
+
+    # symbols is merged into
+    # TwitterCldr::DataReaders::NumberDataReader.new(locale).symbols
+    def localize_maths(f, locale)
+      f.xpath(".//m:mn", MATHML).each do |x|
+        num = /\./.match(x.text) ? x.text.to_f : x.text.to_i
+        precision = /\./.match(x.text) ? x.text.sub(/^.*\./, "").size : 0
+        x.children = localized_number(num, locale, precision)
+      end
+    end
+
+    # By itself twiiter cldr does not support fraction part digits grouping
+    # and custom delimeter, will decorate fraction part manually
+    def localized_number(num, locale, precision)
+      localized = precision == 0 ? num.localize(locale).to_s :
+        num.localize(locale).to_decimal.to_s(:precision => precision)
+      twitter_cldr_reader_symbols = twitter_cldr_reader(locale)
+      return localized unless twitter_cldr_reader_symbols[:decimal]
+      integer, fraction = localized.split(twitter_cldr_reader_symbols[:decimal])
+      return localized if fraction.nil? || fraction.length.zero?
+      [integer, decorate_fraction_part(fraction, locale)].
+        join(twitter_cldr_reader_symbols[:decimal])
+    end
+
+    def decorate_fraction_part(fract, locale)
+      result = []
+      twitter_cldr_reader_symbols = twitter_cldr_reader(locale)
+      fract = fract.slice(0..(twitter_cldr_reader_symbols[:precision] || -1))
+      fr_group_digits = twitter_cldr_reader_symbols[:fraction_group_digits] || 1
+      until fract.empty?
+        result.push(fract.slice!(0, fr_group_digits))
+      end
+      result.join(twitter_cldr_reader_symbols[:fraction_group].to_s)
+    end
+
+    def twitter_cldr_localiser_symbols
+      {}
+    end
+
+    def twitter_cldr_reader(locale)
+      num = TwitterCldr::DataReaders::NumberDataReader.new(locale)
+      num.symbols.merge!(twitter_cldr_localiser_symbols)
+    end
+
+    def twitter_cldr_localiser()
+      locale = TwitterCldr.supported_locale?(@lang.to_sym) ? @lang.to_sym : :en
+      twitter_cldr_reader(locale)
+      locale
+    end
+
+    def mathml1(f, locale)
+      localize_maths(f, locale)
+      return unless f.elements.size == 1 && f.elements.first.name == "mn"
+      f.replace(f.at("./m:mn", MATHML).children)
+    end
+
+    def variant(docxml)
+      docxml.xpath(ns("//variant")).each { |f| variant1(f) }
+      docxml.xpath(ns("//variant[@remove = 'true']")).each { |f| f.remove }
+      docxml.xpath(ns("//variant")).each do |v|
+        next unless v&.next&.name == "variant"
+        v.next = "/"
+      end
+      docxml.xpath(ns("//variant")).each { |f| f.replace(f.children) }
+    end
+
+    def variant1(node)
+      if (!node["lang"] || node["lang"] == @lang) &&
+          (!node["script"] || node["script"] == @script)
+      elsif found_matching_variant_sibling(node)
+        node["remove"] = "true"
+      else
+        #return unless !node.at("./preceding-sibling::xmlns:variant")
+      end
+    end
+
+    def found_matching_variant_sibling(node)
+      prev = node.xpath("./preceding-sibling::xmlns:variant")
+      foll = node.xpath("./following-sibling::xmlns:variant")
+      found = false
+      (prev + foll).each do |n|
+        found = true if n["lang"] == @lang &&
+          (!n["script"] || n["script"] == @script)
+      end
+      found
+    end
   end
 end

data/lib/isodoc/presentation_function/section.rb

@@ -42,5 +42,14 @@ module IsoDoc
       lbl = @xrefs.get[f["id"]][:label] or return
       prefix_name(f, "", "#{lbl}#{clausedelim}", "name")
     end
+
+    def references(docxml)
+    end
+
+    def index(docxml)
+      docxml.xpath(ns("//index | //index-xref")).each do |f|
+        f.remove
+      end
+    end
   end
 end

data/lib/isodoc/presentation_xml_convert.rb

@@ -30,6 +30,8 @@ module IsoDoc
       clause docxml
       annex docxml
       term docxml
+      references docxml
+      index docxml
     end
 
     def block(docxml)
@@ -51,7 +53,10 @@ module IsoDoc
       xref docxml
       eref docxml
       origin docxml
+      concept docxml
       quotesource docxml
+      mathml docxml
+      variant docxml
     end
 
     def postprocess(result, filename, dir)

data/lib/isodoc/version.rb

@@ -1,3 +1,3 @@
 module IsoDoc
-  VERSION = "1.2.6".freeze
+  VERSION = "1.4.0".freeze
 end

data/lib/isodoc/word_convert.rb

@@ -4,26 +4,6 @@ require_relative "word_function/body.rb"
 require_relative "word_function/postprocess.rb"
 
 module IsoDoc
-
-=begin
-  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
-        eval File.open(File.join(File.dirname(__FILE__),"postprocess.rb")).read
-      end
-    end
-  end
-=end
-
   class WordConvert < ::IsoDoc::Convert
     include WordFunction::Comments
     include WordFunction::Footnotes

data/lib/isodoc/word_function/body.rb

@@ -196,6 +196,18 @@ module IsoDoc::WordFunction
       out.parent.at("./table")["class"] = "formula_dl"
     end
 
+    def formula_parse1(node, out)
+      out.div **attr_code(class: "formula") do |div|
+        div.p do |p|
+          parse(node.at(ns("./stem")), div)
+          insert_tab(div, 1)
+          if lbl = node&.at(ns("./name"))&.text
+            div << "(#{lbl})"
+          end
+        end
+      end
+    end
+
     def li_parse(node, out)
       out.li  **attr_code(id: node["id"]) do |li|
         if node["uncheckedcheckbox"] == "true"

data/lib/isodoc/word_function/postprocess.rb

@@ -5,11 +5,9 @@ module IsoDoc::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">
+    <!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: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>
@@ -18,15 +16,13 @@ xmlns:m="http://schemas.microsoft.com/office/2004/12/omml">
 
     def to_word_xhtml_fragment(xml)
       doc = ::Nokogiri::XML.parse(WORD_NOKOHEAD)
-      fragment = ::Nokogiri::XML::DocumentFragment.new(doc, xml, doc.root)
-      fragment
+      ::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|
+      docxml.xpath("//p[not(self::*[@class])][ancestor::*[@class = 'Note']]").each do |p|
         p["class"] = "Note"
       end
     end
@@ -56,8 +52,7 @@ xmlns:m="http://schemas.microsoft.com/office/2004/12/omml">
 
     def word_admonition_images(docxml)
       docxml.xpath("//div[@class = 'Admonition']//img").each do |i|
-        i["width"], i["height"] =
-          Html2Doc.image_resize(i, image_localfile(i), @maxheight, 300)
+        i["width"], i["height"] = Html2Doc.image_resize(i, image_localfile(i), @maxheight, 300)
       end
     end
 
@@ -65,6 +60,7 @@ xmlns:m="http://schemas.microsoft.com/office/2004/12/omml">
       word_annex_cleanup(docxml)
       word_preface(docxml)
       word_nested_tables(docxml)
+      word_colgroup(docxml)
       word_table_align(docxml)
       word_table_separator(docxml)
       word_admonition_images(docxml)
@@ -78,28 +74,44 @@ xmlns:m="http://schemas.microsoft.com/office/2004/12/omml">
       docxml
     end
 
-    def word_nested_tables(docxml)
-      docxml.xpath("//table").each do |t|
-        t.xpath(".//table").reverse.each do |tt|
-          t.next = tt.remove
+    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] do
+              x += 1 
+            end
+            for y2 in r..(r + rs - 1)
+              for x2 in x..(x + cs - 1)
+                cells2d[y2][x2] = 1
+              end
+            end
+            width = (x..(x+cs-1)).each_with_object({width: 0}) { |z, m| m[:width] += w[z] }
+            td["width"] = "#{width[:width]}%"
+            x += cs
+          end
         end
       end
     end
 
-    def authority_cleanup1(docxml, klass)
-      dest = docxml.at("//div[@id = 'boilerplate-#{klass}-destination']")
-      auth = docxml.at("//div[@id = 'boilerplate-#{klass}' or @class = 'boilerplate-#{klass}']")
-      auth&.xpath(".//h1[not(text())] | .//h2[not(text())]")&.each { |h| h.remove }
-      auth&.xpath(".//h1 | .//h2")&.each do |h|
-        h.name = "p"
-        h["class"] = "TitlePageSubhead"
+    # assume percentages
+    def colgroup_widths(t)
+      t.xpath("./colgroup/col").each_with_object([]) do |c, m|
+        m << c["width"].sub(/%$/, "").to_f
       end
-      dest and auth and dest.replace(auth.remove)
     end
 
-    def authority_cleanup(docxml)
-      %w(copyright license legal feedback).each do |t|
-        authority_cleanup1(docxml, t)
+    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
 
@@ -144,19 +156,6 @@ xmlns:m="http://schemas.microsoft.com/office/2004/12/omml">
       end
     end
 
-=begin
-    EMPTY_PARA = "<p style='margin-top:0cm;margin-right:0cm;"\
-      "margin-bottom:0cm;margin-left:0.0pt;margin-bottom:.0001pt;"\
-      "line-height:1.0pt;mso-line-height-rule:exactly'>"\
-      "<span lang=EN-GB style='display:none;mso-hide:all'>&nbsp;</span></p>"
-
-    def table_after_table(docxml)
-     docxml.xpath("//table[following-sibling::*[1]/self::table]").each do |t|
-        t.add_next_sibling(EMPTY_PARA)
-      end
-    end
-=end
-
     def word_table_separator(docxml)
       docxml.xpath("//p[@class = 'TableTitle']").each do |t|
         next unless t.children.empty?
@@ -180,46 +179,6 @@ xmlns:m="http://schemas.microsoft.com/office/2004/12/omml">
       end
     end
 
-    def generate_header(filename, _dir)
-      return nil unless @header
-      template = IsoDoc::Common.liquid(File.read(@header, encoding: "UTF-8"))
-      meta = @meta.get
-      meta[:filename] = filename
-      params = meta.map { |k, v| [k.to_s, v] }.to_h
-      Tempfile.open(%w(header html), :encoding => "utf-8") do |f|
-        f.write(template.render(params))
-        f
-      end
-    end
-
-    def word_section_breaks(docxml)
-      @landscapestyle = ""
-      word_section_breaks1(docxml, "WordSection2")
-      word_section_breaks1(docxml, "WordSection3")
-      word_remove_pb_before_annex(docxml)
-      docxml.xpath("//br[@orientation]").each { |br| br.delete("orientation") }
-    end
-
-    def word_section_breaks1(docxml, sect)
-      docxml.xpath("//div[@class = '#{sect}']//br[@orientation]").reverse.
-        each_with_index do |br, i|
-        @landscapestyle += "\ndiv.#{sect}_#{i} {page:#{sect}"\
-          "#{br["orientation"] == "landscape" ? "L" : "P"};}\n"
-        split_at_section_break(docxml, sect, br, i)
-      end
-    end
-
-    def split_at_section_break(docxml, sect, br, i)
-      move = br.parent.xpath("following::node()") &
-        br.document.xpath("//div[@class = '#{sect}']//*")
-      ins = docxml.at("//div[@class = '#{sect}']").
-        after("<div class='#{sect}_#{i}'/>").next_element
-      move.each do |m|
-        next if m.at("./ancestor::div[@class = '#{sect}_#{i}']")
-        ins << m.remove
-      end
-    end
-
     # applies for <div class="WordSectionN_M"><p><pagebreak/></p>...
     def word_remove_pb_before_annex(docxml)
       docxml.xpath("//div[p/br]").each do |d|
@@ -237,8 +196,7 @@ xmlns:m="http://schemas.microsoft.com/office/2004/12/omml">
       docxml.xpath("//a[@epub:type = 'footnote']").each do |x|
         footnote_reference_format(x)
       end
-      docxml.xpath("//a[@class = 'TableFootnoteRef'] | "\
-                   "//span[@class = 'TableFootnoteRef']").each do |x|
+      docxml.xpath("//a[@class = 'TableFootnoteRef'] | //span[@class = 'TableFootnoteRef']").each do |x|
         table_footnote_reference_format(x)
       end
       docxml