isodoc 0.0.1

data/lib/isodoc/table.rb

@@ -0,0 +1,99 @@
+module IsoDoc
+  class Convert
+    def table_title_parse(node, out)
+      name = node.at(ns("./name"))
+      if name
+        out.p **{ class: "TableTitle", align: "center" } do |p|
+          p.b do |b|
+            b << "#{get_anchors()[node['id']][:label]}&nbsp;&mdash; "
+            b << name.text
+          end
+        end
+      end
+    end
+
+    def thead_parse(node, t)
+      thead = node.at(ns("./thead"))
+      if thead
+        t.thead do |h|
+          thead.element_children.each_with_index do |n, i|
+            tr_parse(n, h, i, thead.element_children.size, true)
+          end
+        end
+      end
+    end
+
+    def tbody_parse(node, t)
+      tbody = node.at(ns("./tbody"))
+      t.tbody do |h|
+        tbody.element_children.each_with_index do |n, i|
+          tr_parse(n, h, i, tbody.element_children.size, false)
+        end
+      end
+    end
+
+    def tfoot_parse(node, t)
+      tfoot = node.at(ns("./tfoot"))
+      if tfoot
+        t.tfoot do |h|
+          tfoot.element_children.each_with_index do |n, i|
+            tr_parse(n, h, i, tfoot.element_children.size, false)
+          end
+        end
+      end
+    end
+
+    def make_table_attr(node)
+      {
+        id: node["id"],
+        class: "MsoISOTable",
+        border: 1,
+        cellspacing: 0,
+        cellpadding: 0,
+      }
+    end
+
+    def table_parse(node, out)
+      @in_table = true
+      table_title_parse(node, out)
+      out.table **make_table_attr(node) do |t|
+        thead_parse(node, t)
+        tbody_parse(node, t)
+        tfoot_parse(node, t)
+        dl = node.at(ns("./dl")) and parse(dl, out)
+        node.xpath(ns("./note")).each { |n| parse(n, out) }
+      end
+      @in_table = false
+      # out.p { |p| p << "&nbsp;" }
+    end
+
+    SW = "solid windowtext"
+
+    #border-left:#{col.zero? ? "#{SW} 1.5pt;" : "none;"}
+    #border-right:#{SW} #{col == totalcols && !header ? "1.5" : "1.0"}pt;
+    def make_tr_attr(td, row, totalrows, col, totalcols, header)
+      style = td.name == "th" ? "font-weight:bold;" : ""
+      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/, "") }
+    end
+
+    def tr_parse(node, out, ord, totalrows, header)
+      out.tr do |r|
+        node.elements.each_with_index do |td, i|
+          attrs = make_tr_attr(td, ord, totalrows - 1, 
+                               i, node.elements.size - 1, header)
+          r.send td.name, **attr_code(attrs) do |entry|
+            td.children.each { |n| parse(n, entry) }
+          end
+        end
+      end
+    end
+  end
+end

data/lib/isodoc/terms.rb

@@ -0,0 +1,74 @@
+module IsoDoc
+  class Convert
+
+    def definition_parse(node, out)
+      node.children.each { |n| parse(n, out) }
+    end
+
+    def modification_parse(node, out)
+      out << "[MODIFICATION]"
+      para = node.at(ns("./p"))
+      para.children.each { |n| parse(n, out) }
+    end
+
+    def deprecated_term_parse(node, out)
+      out.p **{ class: "AltTerms" } do |p|
+        p << "DEPRECATED: #{node.text}"
+      end
+    end
+
+    def admitted_term_parse(node, out)
+      out.p **{ class: "AltTerms" } { |p| p << node.text }
+    end
+
+    def term_parse(node, out)
+      out.p **{ class: "Terms" } { |p| p << node.text }
+    end
+
+    def para_then_remainder(first, node, p)
+      if first.name == "p"
+        first.children.each { |n| parse(n, p) }
+        node.elements.drop(1).each { |n| parse(n, div) }
+      else
+        node.elements.each { |n| parse(n, div) }
+      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 << "EXAMPLE:"
+          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)
+        end
+      end
+    end
+
+    def termref_parse(node, out)
+      out.p do |p|
+        p << "[TERMREF]"
+        node.children.each { |n| parse(n, p) }
+        p << "[/TERMREF]"
+      end
+    end
+
+    def termdef_parse(node, out)
+      out.p **{ class: "TermNum", id: node["id"] } do |p|
+        p << get_anchors()[node["id"]][:label]
+      end
+      set_termdomain("")
+      node.children.each { |n| parse(n, out) }
+    end
+  end
+end

data/lib/isodoc/utils.rb

@@ -0,0 +1,88 @@
+#require "uuidtools"
+
+module IsoDoc
+  class Convert
+    def ns(xpath)
+      xpath.gsub(%r{/([a-zA-z])}, "/xmlns:\\1").
+        gsub(%r{::([a-zA-z])}, "::xmlns:\\1").
+        gsub(%r{\[([a-zA-z]+ ?=)}, "[xmlns:\\1").
+        gsub(%r{\[([a-zA-z]+\])}, "[xmlns:\\1")
+    end
+
+    def insert_tab(out, n)
+      out.span **attr_code(style: "mso-tab-count:#{n}") do |span|
+        [1..n].each { |i| span << "&#xA0; " }
+      end
+    end
+
+    STAGE_ABBRS = {
+      "00": "PWI",
+      "10": "NWIP",
+      "20": "WD",
+      "30": "CD",
+      "40": "DIS",
+      "50": "FDIS",
+      "60": "IS",
+      "90": "(Review)",
+      "95": "(Withdrawal)",
+    }.freeze
+
+    def stage_abbreviation(stage)
+      STAGE_ABBRS[stage.to_sym] || "??"
+    end
+
+    NOKOHEAD = <<~HERE
+    <!DOCTYPE html SYSTEM
+    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
+    <html xmlns="http://www.w3.org/1999/xhtml">
+    <head> <title></title> <meta charset="UTF-8" /> </head>
+    <body> </body> </html>
+    HERE
+
+
+    # block for processing XML document fragments as XHTML,
+    # to allow for HTMLentities
+    def noko(&block)
+      doc = ::Nokogiri::XML.parse(NOKOHEAD)
+      fragment = doc.fragment("")
+      ::Nokogiri::XML::Builder.with fragment, &block
+      fragment.to_xml(encoding: "US-ASCII").lines.map do |l|
+        l.gsub(/\s*\n/, "")
+      end
+    end
+
+    def attr_code(attributes)
+      attributes = attributes.reject { |_, val| val.nil? }.map
+      attributes.map do |k, v|
+        [k, (v.is_a? String) ? HTMLEntities.new.decode(v) : v]
+      end.to_h
+    end
+
+    NOKOHEAD = <<~HERE
+          <!DOCTYPE html SYSTEM
+          "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
+          <html xmlns="http://www.w3.org/1999/xhtml">
+          <head> <title></title> <meta charset="UTF-8" /> </head>
+          <body> </body> </html>
+    HERE
+
+    def to_xhtml(xml)
+      xml.gsub!(/<\?xml[^>]*>/, "")
+      unless /<!DOCTYPE /.match? xml
+        xml = '<!DOCTYPE html SYSTEM
+          "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">' + xml
+      end
+      Nokogiri::XML.parse(xml)
+    end
+
+    def to_xhtml_fragment(xml)
+      doc = ::Nokogiri::XML.parse(NOKOHEAD)
+      fragment = doc.fragment(xml)
+      fragment
+    end
+
+    def from_xhtml(xml)
+      xml.to_xml.sub(%r{ xmlns="http://www.w3.org/1999/xhtml"}, "")
+    end
+  end
+end

data/lib/isodoc/version.rb

@@ -0,0 +1,3 @@
+module IsoDoc
+  VERSION = "0.0.1".freeze
+end

data/lib/isodoc/xref_gen.rb

@@ -0,0 +1,204 @@
+#require "isodoc/utils"
+
+module IsoDoc
+  class Convert
+    #include ::IsoDoc::Utils
+
+    @anchors = {}
+
+    def get_anchors
+      @anchors
+    end
+
+    def back_anchor_names(docxml)
+      docxml.xpath(ns("//annex")).each_with_index do |c, i|
+        annex_names(c, (65 + i).chr.to_s)
+      end
+      docxml.xpath(ns("//bibitem")).each do |ref|
+        reference_names(ref)
+      end
+    end
+
+    def initial_anchor_names(d)
+      introduction_names(d.at(ns("//content[title = 'Introduction']")))
+      section_names(d.at(ns("//clause[title = 'Scope']")), "1", 1)
+      section_names(d.at(ns(
+        "//references[title = 'Normative References']")), "2", 1)
+      section_names(d.at(ns("//terms")), "3", 1)
+      middle_section_asset_names(d)
+    end
+
+    def middle_section_asset_names(d)
+      middle_sections = "//clause[title = 'Scope'] | "\
+        "//references[title = 'Normative References'] | //terms | "\
+        "//symbols-abbrevs | //clause[parent::sections]"
+      sequential_asset_names(d.xpath(ns(middle_sections)))
+    end
+
+    def clause_names(docxml,sect_num)
+      q = "//clause[parent::sections][not(xmlns:title = 'Scope')]"
+      docxml.xpath(ns(q)).each_with_index do |c, i|
+        section_names(c, (i + sect_num).to_s, 1)
+      end
+    end
+
+    def termnote_anchor_names(docxml)
+      docxml.xpath(ns("//term[termnote]")).each do |t|
+        t.xpath(ns("./termnote")).each_with_index do |n, i|
+          @anchors[n["id"]] = { label: "Note #{i + 1} to entry",
+                                xref: "#{@anchors[t["id"]][:xref]},"\
+                                "Note #{i + 1}" }
+        end
+      end
+    end
+
+    def table_note_anchor_names(docxml)
+      docxml.xpath(ns("//table[note]")).each do |t|
+        t.xpath(ns("./note")).each_with_index do |n, i|
+          @anchors[n["id"]] = { label: "NOTE #{i + 1}",
+                                xref: "#{@anchors[t["id"]][:xref]},"\
+                                "Note #{i + 1}" }
+        end
+      end
+    end
+
+    def middle_anchor_names(docxml)
+      symbols_abbrevs = docxml.at(ns("//symbols-abbrevs"))
+      sect_num = 4
+      if symbols_abbrevs
+        section_names(symbols_abbrevs, sect_num.to_s, 1)
+        sect_num += 1
+      end
+      clause_names(docxml, sect_num)
+      termnote_anchor_names(docxml)
+    end
+
+    # extract names for all anchors, xref and label
+    def anchor_names(docxml)
+      initial_anchor_names(docxml)
+      middle_anchor_names(docxml)
+      back_anchor_names(docxml)
+      table_note_anchor_names(docxml)
+    end
+
+    def sequential_figure_names(clause)
+      i = j = 0
+      clause.xpath(ns(".//figure")).each do |t|
+        label = "Figure #{i}" + ( j.zero? ? "" : "-#{j}" )
+        if t.parent.name == "figure"
+          j += 1
+        else
+          j = 0
+          i += 1
+        end
+        label = "Figure #{i}" + ( j.zero? ? "" : "-#{j}" )
+        @anchors[t["id"]] = { label: label, xref: label }
+      end
+    end
+
+    def sequential_asset_names(clause)
+      clause.xpath(ns(".//table")).each_with_index do |t, i|
+        @anchors[t["id"]] = { label: "Table #{i + 1}",
+                              xref: "Table #{i + 1}" }
+      end
+      sequential_figure_names(clause)
+      clause.xpath(ns(".//formula")).each_with_index do |t, i|
+        @anchors[t["id"]] = { label: (i + 1).to_s,
+                              xref: "Formula #{i + 1}" }
+      end
+    end
+
+    def hierarchical_figure_names(clause, num)
+      i = j = 0
+      clause.xpath(ns(".//figure")).each do |t|
+        if t.parent.name == "figure"
+          j += 1
+        else
+          j = 0
+          i += 1
+        end
+        label = "Figure #{num}.#{i}" + ( j.zero? ? "" : "-#{j}" )
+        @anchors[t["id"]] = { label: label, xref: label }
+      end
+    end
+
+    def hierarchical_asset_names(clause, num)
+      clause.xpath(ns(".//table")).each_with_index do |t, i|
+        @anchors[t["id"]] = { label: "Table #{num}.#{i + 1}",
+                              xref: "Table #{num}.#{i + 1}" }
+      end
+      hierarchical_figure_names(clause, num)
+      clause.xpath(ns(".//formula")).each_with_index do |t, i|
+        @anchors[t["id"]] = { label: "#{num}.#{i + 1}",
+                              xref: "Formula #{num}.#{i + 1}" }
+      end
+    end
+
+    def introduction_names(clause)
+      clause.xpath(ns("./subsection")).each_with_index do |c, i|
+        section_names(c, "0.#{i + 1}")
+      end
+    end
+
+    def section_names(clause, num, level)
+      @anchors[clause["id"]] = { label: num, xref: "Clause #{num}",
+                                 level: level }
+      clause.xpath(ns("./subsection | ./term")).each_with_index do |c, i|
+        section_names1(c, "#{num}.#{i + 1}", level + 1)
+      end
+    end
+
+    def section_names1(clause, num, level)
+      @anchors[clause["id"]] = 
+        { label: num, level: level,
+          xref: clause.name == "term" ? num : "Clause #{num}" }
+      clause.xpath(ns("./subsection ")).
+        each_with_index do |c, i|
+        section_names1(c, "#{num}.#{i + 1}", level + 1)
+      end
+    end
+
+    def annex_names(clause, num)
+      obligation = "(Informative)"
+      obligation = "(Normative)" if clause["subtype"] == "normative"
+      label = "<b>Annex #{num}</b><br/>#{obligation}"
+      @anchors[clause["id"]] = { label: label,
+                                 xref: "Annex #{num}", level: 1 }
+      clause.xpath(ns("./subsection")).each_with_index do |c, i|
+        annex_names1(c, "#{num}.#{i + 1}", 2)
+      end
+      hierarchical_asset_names(clause, num)
+    end
+
+    def annex_names1(clause, num, level)
+      @anchors[clause["id"]] = { label: num,
+                                 xref: num,
+                                 level: level }
+      clause.xpath(ns(".//subsection")).each_with_index do |c, i|
+        annex_names1(c, "#{num}.#{i + 1}", level + 1)
+      end
+    end
+
+    def format_ref(ref, isopub)
+      return "ISO #{ref}" if isopub
+      return "[#{ref}]" if /^\d+$/.match?(ref) && !/^\[.*\]$/.match?(ref)
+      ref
+    end
+
+    def reference_names(ref)
+      isopub = ref.at(ns("./publisher/affiliation[name = 'ISO']"))
+      docid = ref.at(ns("./docidentifier"))
+      return ref_names(ref) unless docid
+      date = ref.at(ns("./publisherdate"))
+      reference = format_ref(docid.text, isopub)
+      reference += ": #{date.text}" if date && isopub
+      @anchors[ref["id"]] = { xref: reference }
+    end
+
+    def ref_names(ref)
+      linkend = ref.text
+      linkend.gsub!(/[\[\]]/, "") unless /^\[\d+\]$/.match? linkend
+      @anchors[ref["id"]] = { xref: linkend }
+    end
+  end
+end