isodoc 0.0.1

data/lib/isodoc/metadata.rb

@@ -0,0 +1,99 @@
+require "htmlentities"
+
+module IsoDoc
+  class Convert
+
+    def get_metadata
+      @meta
+    end
+
+    def set_metadata(key, value)
+      @meta[key] = value
+    end
+
+    def author(isoxml, _out)
+      # tc = isoxml.at(ns("//technical-committee"))
+      tc_num = isoxml.at(ns("//technical-committee/@number"))
+      # sc = isoxml.at(ns("//subcommittee"))
+      sc_num = isoxml.at(ns("//subcommittee/@number"))
+      # wg = isoxml.at(ns("//workgroup"))
+      wg_num = isoxml.at(ns("//workgroup/@number"))
+      secretariat = isoxml.at(ns("//secretariat"))
+      set_metadata(:tc, "XXXX")
+      set_metadata(:sc, "XXXX")
+      set_metadata(:wg, "XXXX")
+      set_metadata(:secretariat, "XXXX")
+      set_metadata(:tc,  tc_num.text) if tc_num
+      set_metadata(:sc, sc_num.text) if sc_num
+      set_metadata(:wg, wg_num.text) if wg_num
+      set_metadata(:secretariat, secretariat.text) if secretariat
+    end
+
+    def id(isoxml, _out)
+      docnumber = isoxml.at(ns("//project-number"))
+      partnumber = isoxml.at(ns("//project-number/@part"))
+      documentstatus = isoxml.at(ns("//status/stage"))
+      dn = docnumber.text
+      dn += "-#{partnumber.text}" if partnumber
+      if documentstatus
+        set_metadata(:stage, documentstatus.text)
+        abbr = stage_abbreviation(documentstatus.text)
+        set_metadata(:stageabbr, abbr)
+        documentstatus.text.to_i < 60 and
+          dn = abbr + " " + dn
+      end
+      set_metadata(:docnumber, dn)
+    end
+
+    def draftinfo(draft, revdate)
+      draftinfo = ""
+      if draft
+        draftinfo = " (draft #{draft.text}"
+        draftinfo += ", #{revdate.text}" if revdate
+        draftinfo += ")"
+      end
+      draftinfo
+    end
+
+    def version(isoxml, _out)
+      yr = isoxml.at(ns("//copyright/from"))
+      set_metadata(:docyear, yr.text)
+      draft = isoxml.at(ns("//version/draft"))
+      set_metadata(:draft, draft.nil? ? nil : draft.text)
+      revdate = isoxml.at(ns("//version/revision-date"))
+      set_metadata(:revdate, revdate.nil? ? nil : revdate.text)
+      draftinfo = draftinfo(draft, revdate)
+      set_metadata(:draftinfo, draftinfo(draft, revdate))
+    end
+
+
+    def compose_title(main, intro, part, partnumber)
+      c = HTMLEntities.new
+      main = c.encode(main.text, :hexadecimal)
+      intro &&
+        main = "#{c.encode(intro.text, :hexadecimal)}&nbsp;&mdash; #{main}"
+      part &&
+        main = "#{main}&nbsp;&mdash; Part&nbsp;#{partnumber}: "\
+        "#{c.encode(part.text, :hexadecimal)}"
+      main
+    end
+
+    def title(isoxml, _out)
+      intro = isoxml.at(ns("//title[@language='en']/title-intro"))
+      main = isoxml.at(ns("//title[@language='en']/title-main"))
+      part = isoxml.at(ns("//title[@language='en']/title-part"))
+      partnumber = isoxml.at(ns("//id/project-number/@part"))
+      main = compose_title(main, intro, part, partnumber)
+      set_metadata(:doctitle, main)
+    end
+
+    def subtitle(isoxml, _out)
+      intro = isoxml.at(ns("//title[@language='fr']/title-intro"))
+      main = isoxml.at(ns("//title[@language='fr']/title-main"))
+      part = isoxml.at(ns("//title[@language='fr']/title-part"))
+      partnumber = isoxml.at(ns("//id/project-number/@part"))
+      main = compose_title(main, intro, part, partnumber)
+      set_metadata(:docsubtitle, main)
+    end
+  end
+end

data/lib/isodoc/postprocessing.rb

@@ -0,0 +1,156 @@
+require "html2doc"
+require "htmlentities"
+require "nokogiri"
+require "pp"
+
+module IsoDoc
+  class Convert
+
+    def postprocess(result, filename, dir)
+      generate_header(filename, dir)
+      result = from_xhtml(cleanup(to_xhtml(result)))
+      toWord(result, filename, dir)
+      toHTML(result, filename)
+    end
+
+    def toWord(result, filename, dir)
+      result = from_xhtml(wordCleanup(to_xhtml(result)))
+      result = populate_template(result)
+      Html2Doc.process(result, filename, @wordstylesheet, "header.html", 
+                       dir, ['`', '`'])
+    end
+
+    def wordCleanup(docxml)
+      wordPreface(docxml)
+      wordAnnexCleanup(docxml)
+      docxml
+    end
+
+    # force Annex h2 to be p.h2Annex, so it is not picked up by ToC
+    def wordAnnexCleanup(docxml)
+      d = docxml.xpath("//h2[ancestor::*[@class = 'Section3']]").each do |h2|
+        h2.name = "p"
+        h2["class"] = "h2Annex"
+      end
+    end
+
+    def wordPreface(docxml)
+      cover = to_xhtml_fragment(File.read(@wordcoverpage, encoding: "UTF-8"))
+      d = docxml.at('//div[@class="WordSection1"]')
+      d.children.first.add_previous_sibling cover.to_xml(encoding: 'US-ASCII')
+      intro = to_xhtml_fragment(
+        File.read(@wordintropage, encoding: "UTF-8").
+        sub(/WORDTOC/, makeWordToC(docxml)))
+      d = docxml.at('//div[@class="WordSection2"]')
+      d.children.first.add_previous_sibling intro.to_xml(encoding: 'US-ASCII')
+    end
+
+    def populate_template(docxml)
+      meta = get_metadata
+      docxml.
+        gsub(/DOCYEAR/, meta[:docyear]).
+        gsub(/DOCNUMBER/, meta[:docnumber]).
+        gsub(/TCNUM/, meta[:tc]).
+        gsub(/SCNUM/, meta[:sc]).
+        gsub(/WGNUM/, meta[:wg]).
+        gsub(/DOCTITLE/, meta[:doctitle]).
+        gsub(/DOCSUBTITLE/, meta[:docsubtitle]).
+        gsub(/SECRETARIAT/, meta[:secretariat]).
+        gsub(/[ ]?DRAFTINFO/, meta[:draftinfo]).
+        gsub(/\[TERMREF\]\s*/, "[SOURCE: ").
+        gsub(/\s*\[\/TERMREF\]\s*/, "]").
+        gsub(/\s*\[ISOSECTION\]/, ", ").
+        gsub(/\s*\[MODIFICATION\]/, ", modified — ").
+        gsub(%r{WD/CD/DIS/FDIS}, meta[:stageabbr])
+    end
+
+    def generate_header(filename, dir)
+      header = File.read(@header, encoding: "UTF-8").
+        gsub(/FILENAME/, filename).
+        gsub(/DOCYEAR/, get_metadata()[:docyear]).
+        gsub(/[ ]?DRAFTINFO/, get_metadata()[:draftinfo]).
+        gsub(/DOCNUMBER/, get_metadata()[:docnumber])
+      File.open("header.html", "w") do |f|
+        f.write(header)
+      end
+    end
+
+    # these are in fact preprocess,
+    # but they are extraneous to main HTML file
+    def html_header(html, docxml, filename, dir)
+      anchor_names docxml
+      define_head html, filename, dir
+    end
+
+    # isodoc.css overrides any CSS injected by Html2Doc, which
+    # is inserted before this CSS.
+    def define_head(html, filename, dir)
+      html.head do |head|
+        head.title { |t| t << filename }
+        head.style do |style|
+          stylesheet = File.read(@standardstylesheet).
+            gsub("FILENAME", filename)
+          style.comment "\n#{stylesheet}\n"
+        end
+      end
+    end
+
+    def titlepage(_docxml, div)
+      titlepage = File.read(@wordcoverpage, encoding: "UTF-8")
+      div.parent.add_child titlepage
+    end
+
+    def wordTocEntry(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
+      <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
+      <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 header_strip(h)
+      h.to_s.gsub(%r{<br/>}, " ").
+        sub(/<h[12][^>]*>/, "").sub(%r{</h[12]>}, "")
+    end
+
+    def makeWordToC(docxml)
+      toc = ""
+      docxml.xpath("//h1 | //h2[not(ancestor::*[@class = 'Section3'])]").
+        each do |h|
+        toc += wordTocEntry(h.name == "h1" ? 1 : 2, header_strip(h))
+      end
+      toc.sub(/(<p class="MsoToc1">)/, 
+              %{\\1#{WORD_TOC_PREFACE}}) + WORD_TOC_SUFFIX
+    end
+
+  end
+end

data/lib/isodoc/references.rb

@@ -0,0 +1,129 @@
+module IsoDoc
+  class Convert
+    def iso_bibitem_ref_code(b)
+      isocode = b.at(ns("./docidentifier"))
+      isodate = b.at(ns("./publishdate"))
+      reference = "ISO #{isocode.text}"
+      reference += ": #{isodate.text}" if isodate
+      reference
+    end
+
+    def date_note_process(b, ref)
+      date_note = b.xpath(ns("./note[text()][contains(.,'ISO DATE:')]"))
+      unless date_note.empty?
+        date_note.first.content =
+          date_note.first.content.gsub(/ISO DATE: /, "")
+        date_note.wrap("<p></p>")
+        footnote_parse(date_note.first, ref)
+      end
+    end
+
+    def iso_bibitem_entry(list, b, ordinal, biblio)
+      attrs = { id: b["id"], class: biblio ? "Biblio" : nil }
+      list.p **attr_code(attrs) do |ref|
+        if biblio
+          ref << "[#{ordinal}]"
+          insert_tab(ref, 1)
+        end
+        ref << iso_bibitem_ref_code(b)
+        date_note_process(b, ref)
+        ref << ", " if biblio
+        ref.i { |i| i << " #{b.at(ns('./name')).text}" }
+      end
+    end
+
+    def ref_entry_code(r, ordinal, t)
+      if /^\d+$/.match?(t)
+        r << "[#{t}]"
+        insert_tab(r, 1)
+      else
+        r << "[#{ordinal}]"
+        insert_tab(r, 1)
+        r << "#{t},"
+      end
+    end
+
+    def ref_entry(list, b, ordinal, bibliography)
+      ref = b.at(ns("./ref"))
+      para = b.at(ns("./p"))
+      list.p **attr_code("id": ref["id"], class: "Biblio") do |r|
+        ref_entry_code(r, ordinal, ref.text.gsub(/[\[\]]/, ""))
+        para.children.each { |n| parse(n, r) }
+      end
+    end
+
+    def noniso_bibitem(list, b, ordinal, bibliography)
+      ref = b.at(ns("./docidentifier"))
+      para = b.at(ns("./formatted"))
+      list.p **attr_code("id": b["id"], class: "Biblio") do |r|
+        ref_entry_code(r, ordinal, ref.text.gsub(/[\[\]]/, ""))
+        para.children.each { |n| parse(n, r) }
+      end
+    end
+
+    def split_bibitems(f)
+      iso_bibitem = []
+      non_iso_bibitem = []
+      f.xpath(ns("./bibitem")).each do |x|
+        if x.at(ns("./publisher/affiliation[name = 'ISO']")).nil?
+          non_iso_bibitem << x
+        else
+          iso_bibitem << x
+        end
+      end
+      { iso: iso_bibitem, noniso: non_iso_bibitem }
+    end
+
+    def biblio_list(f, div, bibliography)
+      bibitems = split_bibitems(f)
+      bibitems[:iso].each_with_index do |b, i|
+        iso_bibitem_entry(div, b, (i + 1), bibliography)
+      end
+      bibitems[:noniso].each_with_index do |b, i|
+        noniso_bibitem(div, b, (i + 1 + bibitems[:iso].size), bibliography)
+      end
+    end
+
+    NORM_WITH_REFS_PREF = <<~BOILERPLATE
+          The following documents are referred to in the text in such a way
+          that some or all of their content constitutes requirements of this
+          document. For dated references, only the edition cited applies.
+          For undated references, the latest edition of the referenced
+          document (including any amendments) applies.
+    BOILERPLATE
+
+    NORM_EMPTY_PREF =
+      "There are no normative references in this document."
+
+    def norm_ref_preface(f, div)
+      refs = f.elements.select do |e|
+        ["reference", "bibitem"].include? e.name
+      end
+      pref = refs.empty? ? NORM_EMPTY_PREF : NORM_WITH_REFS_PREF
+      div.p pref
+    end
+
+    def norm_ref(isoxml, out)
+      q = "//sections/references[title = 'Normative References']"
+      f = isoxml.at(ns(q)) or return
+      out.div do |div|
+        clause_name("2.", "Normative References", div, false)
+        norm_ref_preface(f, div)
+        biblio_list(f, div, false)
+      end
+    end
+
+    def bibliography(isoxml, out)
+      q = "//sections/references[title = 'Bibliography']"
+      f = isoxml.at(ns(q)) or return
+      page_break(out)
+      out.div do |div|
+        div.h1 "Bibliography", **{ class: "Section3" }
+        f.elements.reject do |e|
+          ["reference", "title", "bibitem"].include? e.name
+        end.each { |e| parse(e, div) }
+        biblio_list(f, div, true)
+      end
+    end
+  end
+end

data/lib/isodoc/section.rb

@@ -0,0 +1,136 @@
+module IsoDoc
+  class Convert
+    def clause_parse(node, out)
+      out.div **attr_code("id": node["id"]) do |s|
+        node.children.each do |c1|
+          if c1.name == "title"
+            if node["inline-header"]
+              out.span **{ class: "zzMoveToFollowing" } do |s|
+                s.b do |b|
+                  b << "#{get_anchors()[node['id']][:label]}. #{c1.text} "
+                end
+              end
+            else
+              s.send "h#{get_anchors()[node['id']][:level]}" do |h|
+                h << "#{get_anchors()[node['id']][:label]}. #{c1.text}"
+              end
+            end
+          else
+            parse(c1, s)
+          end
+        end
+      end
+    end
+
+    def clause_name(num, title, div, inline_header)
+      if inline_header
+        div.span **{ class: "zzMoveToFollowing" } do |s|
+          s.b do |b|
+            b << num
+            b << title + " "
+          end
+        end
+      else
+        div.h1 do |h1|
+          h1 << num
+          insert_tab(h1, 1)
+          h1 << title
+        end
+      end
+    end
+
+    def clause(isoxml, out)
+      isoxml.xpath(ns("//clause[parent::sections]")).each do |c|
+        next if c.at(ns("./title")).text == "Scope"
+        out.div **attr_code("id": c["id"]) do |s|
+          c.elements.each do |c1|
+            if c1.name == "title"
+              clause_name("#{get_anchors()[c['id']][:label]}.", 
+                          c1.text, s, c["inline-header"])
+            else
+              parse(c1, s)
+            end
+          end
+        end
+      end
+    end
+
+    def annex_name(annex, name, div)
+      div.h1 **{ class: "Annex" } do |t|
+        t << "#{get_anchors()[annex['id']][:label]}<br/><br/>"
+        t << "<b>#{name.text}</b>"
+      end
+    end
+
+    def annex(isoxml, out)
+      isoxml.xpath(ns("//annex")).each do |c|
+        page_break(out)
+        out.div **attr_code("id": c["id"], class: "Section3" ) do |s|
+          #s1.div **{ class: "annex" } do |s|
+            c.elements.each do |c1|
+              if c1.name == "title" then annex_name(c, c1, s)
+              else
+                parse(c1, s)
+              end
+            end
+          # end
+        end
+      end
+    end
+
+    def scope(isoxml, out)
+      f = isoxml.at(ns("//clause[title = 'Scope']")) || return
+      out.div do |div|
+        clause_name("1.", "Scope", div, false)
+        f.elements.each do |e|
+          parse(e, div) unless e.name == "title"
+        end
+      end
+    end
+
+    def terms_defs(isoxml, out)
+      f = isoxml.at(ns("//terms")) || return
+      out.div do |div|
+        clause_name("3.", "Terms and Definitions", div, false)
+        f.elements.each do |e|
+          parse(e, div) unless e.name == "title"
+        end
+      end
+    end
+
+    def symbols_abbrevs(isoxml, out)
+      f = isoxml.at(ns("//symbols-abbrevs")) || return
+      out.div do |div|
+        clause_name("4.", "Symbols and Abbreviations", div, false)
+        f.elements.each do |e|
+          parse(e, div) unless e.name == "title"
+        end
+      end
+    end
+
+    def introduction(isoxml, out)
+      f = isoxml.at(ns("//content[title = 'Introduction']")) || return
+      title_attr = { class: "IntroTitle" }
+      page_break(out)
+      out.div **{ class: "Section3" } do |div|
+        div.h1 "Introduction", **attr_code(title_attr)
+        f.elements.each do |e|
+          if e.name == "patent-notice"
+            e.elements.each { |e1| parse(e1, div) }
+          else
+            parse(e, div) unless e.name == "title"
+          end
+        end
+      end
+    end
+
+    def foreword(isoxml, out)
+      f = isoxml.at(ns("//content[title = 'Foreword']")) || return
+      page_break(out)
+      out.div do |s|
+        s.h1 **{ class: "ForewordTitle" } { |h1| h1 << "Foreword" }
+        f.elements.each { |e| parse(e, s) unless e.name == "title" }
+      end
+    end
+  end
+end