metanorma-jis 0.0.2 → 0.0.4

data/lib/isodoc/jis/metadata.rb

@@ -49,6 +49,13 @@ module IsoDoc
       end
 
       PART_LABEL = { en: "Part", ja: "その" }.freeze
+
+      def docid(isoxml, _out)
+        id = isoxml.at(ns("//bibdata/docidentifier[@type = 'JIS']"))&.text or
+          return
+        set(:docnumber, id)
+        set(:docnumber_undated, id.sub(/:\d{4}$/, ""))
+      end
     end
   end
 end

data/lib/isodoc/jis/presentation_xml_convert.rb

@@ -4,6 +4,175 @@ require "isodoc"
 module IsoDoc
   module JIS
     class PresentationXMLConvert < IsoDoc::Iso::PresentationXMLConvert
+      def inline(docxml)
+        super
+        strong(docxml)
+      end
+
+      JPAN = "\\p{Hiragana}\\p{Katakana}\\p{Han}".freeze
+      JPAN_BOLD = "<span style='font-family:\"MS Gothic\"'>".freeze
+
+      def strong(docxml)
+        docxml.xpath(ns("//strong")).each do |x|
+          (x.children.size == 1 && x.children.first.text?) or next # too hard
+          x.replace(strong1(x.text))
+        end
+      end
+
+      def strong1(text)
+        jpan = /^[#{JPAN}]/o.match?(text[0])
+        ret = jpan ? JPAN_BOLD : "<strong>"
+        text.chars.each do |n|
+          new = /^[#{JPAN}]/o.match?(n)
+          jpan && !new and ret += "</span><strong>"
+          !jpan && new and ret += "</strong>#{JPAN_BOLD}"
+          ret += n
+          jpan = new
+        end
+        ret += /[#{JPAN}]/o.match?(text[-1]) ? "</span>" : "</strong>"
+        ret
+      end
+
+      def admits(elem)
+        elem.children.first.previous = @i18n.l10n("#{@i18n.admitted}: ")
+      end
+
+      def block(docxml)
+        super
+        dl docxml
+      end
+
+      def dl(docxml)
+        docxml.xpath(ns("//table//dl | //figure//dl")).each do |l|
+          l.at(ns("./dl")) || l.at("./ancestor::xmlns:dl") and next
+          dl_to_para(l)
+        end
+      end
+
+      def dt_dd?(node)
+        %w{dt dd}.include? node.name
+      end
+
+      def dl_to_para(node)
+        ret = dl_to_para_name(node)
+        ret += dl_to_para_terms(node)
+        node.elements.reject { |n| %w(dt dd name).include?(n.name) }.each do |x|
+          ret += x.to_xml
+        end
+        dl_id_insert(node, ret)
+      end
+
+      def dl_id_insert(node, ret)
+        a = node.replace(ret)
+        p = a.at("./descendant-or-self::xmlns:p")
+        node["id"] and p << "<bookmark id='#{node['id']}'/>"
+        a.xpath("./descendant-or-self::*[@id = '']").each { |x| x.delete("id") }
+      end
+
+      def dl_to_para_name(node)
+        e = node.at(ns("./name"))
+        "<p class='ListTitle'>#{e&.children&.to_xml || @i18n.key}</p>"
+      end
+
+      def dl_to_para_terms(node)
+        ret = ""
+        node.elements.select { |n| dt_dd?(n) }.each_slice(2) do |dt, dd|
+          term = strip_para(dt)
+          defn = strip_para(dd)
+          bkmk = dd["id"] ? "<bookmark id='#{dd['id']}'/>" : ""
+          ret += "<p class='dl' id='#{dt['id']}'>#{term}: #{bkmk}#{defn}</p>"
+        end
+        ret
+      end
+
+      def strip_para(node)
+        node.children.to_xml.gsub(%r{</?p( [^>]*)?>}, "")
+      end
+
+      def table1(node)
+        super
+        cols = table_cols_count(node)
+        name = node.at(ns("./name"))
+        thead = table_thead_pt(node, name)
+        table_unit_note(node, thead, cols)
+        table_name(name, thead, cols)
+      end
+
+      def table_thead_pt(node, name)
+        node.at(ns("./thead")) ||
+          name&.after("<thead> </thead>")&.next ||
+          node.elements.first.before("<thead> </thead>").previous
+      end
+
+      def table_cols_count(node)
+        cols = 0
+        node.at(ns(".//tr")).xpath(ns("./td | ./th")).each do |x|
+          cols += x["colspan"]&.to_i || 1
+        end
+        cols
+      end
+
+      def table_unit_note(node, thead, cols)
+        unit_note = node.at(ns(".//note[@type = 'units']")) or return
+        thead.children.first.previous = full_row(cols, unit_note.remove.to_xml)
+      end
+
+      def table_name(name, thead, cols)
+        name or return
+        thead.children.first.previous =
+          full_row(cols, "<p class='TableTitle' style='text-align:center;'> " \
+                         "#{name.remove.children.to_xml}</p>")
+      end
+
+      def full_row(cols, elem)
+        "<tr><td border='0' colspan='#{cols}'>#{elem}</td></tr>"
+      end
+
+      def annex1(elem)
+        elem["commentary"] == "true" and return
+        lbl = @xrefs.anchor(elem["id"], :label)
+        if t = elem.at(ns("./title"))
+          t.children = "<strong>#{to_xml(t.children)}</strong>"
+        end
+        prefix_name(elem, "<br/>", lbl, "title")
+      end
+
+      def annex(docxml)
+        super
+        move_commentaries_to_end(docxml)
+      end
+
+      def move_commentaries_to_end(docxml)
+        docxml.at(ns("//annex[@commentary = 'true']")) or return
+        b = docxml.at(ns("//bibliography")) ||
+          docxml.at(ns("//annex[last()]")).after(" ").next
+        docxml.xpath(ns("//annex[@commentary = 'true']")).reverse.each do |x|
+          b.next = x.remove
+        end
+      end
+
+      def display_order(docxml)
+        i = 0
+        i = display_order_xpath(docxml, "//preface/*", i)
+        i = display_order_at(docxml, "//clause[@type = 'scope']", i)
+        i = display_order_at(docxml, @xrefs.klass.norm_ref_xpath, i)
+        i = display_order_at(docxml, "//sections/terms | " \
+                                     "//sections/clause[descendant::terms]", i)
+        i = display_order_at(docxml, "//sections/definitions", i)
+        i = display_order_xpath(docxml, @xrefs.klass.middle_clause(docxml), i)
+        i = display_order_xpath(docxml, "//annex[not(@commentary = 'true')]", i)
+        i = display_order_xpath(docxml, @xrefs.klass.bibliography_xpath, i)
+        i = display_order_xpath(docxml, "//annex[@commentary = 'true']", i)
+        i = display_order_xpath(docxml, "//indexsect", i)
+        display_order_xpath(docxml, "//colophon/*", i)
+      end
+
+      def tablesource(elem)
+        while elem&.next_element&.name == "source"
+          elem << "; #{to_xml(elem.next_element.remove.children)}"
+        end
+        elem.children = l10n("#{@i18n.source}: #{to_xml(elem.children).strip}")
+      end
 
       include Init
     end

data/lib/isodoc/jis/table.rb

@@ -0,0 +1,58 @@
+module IsoDoc
+  module JIS
+    class WordConvert < IsoDoc::Iso::WordConvert
+      def make_table_footnote_target(out, fnid, fnref)
+        attrs = { id: fnid, class: "TableFootnoteRef" }
+        out.span do |s|
+          s << @i18n.table_footnote
+          out.span **attrs do |a|
+            a << "#{fnref})"
+          end
+          insert_tab(s, 1)
+        end
+      end
+
+      def table_title_parse(node, out); end
+
+      def table_attrs(node)
+        { id: node["id"], title: node["alt"],
+          summary: node["summary"], width: node["width"],
+          class: (node.text.length > 4000 ? "MsoTableGridBig" : "MsoTableGrid"),
+          style: "border-collapse:collapse;" \
+                 "mso-table-anchor-horizontal:column;mso-table-overlap:never;" \
+                 "border:none;mso-padding-alt: " \
+                 "0cm 5.4pt 0cm 5.4pt;mso-border-insideh:none;" \
+                 "mso-border-insidev:none;#{keep_style(node)}",
+          border: 0, cellspacing: 0, cellpadding: 0 }
+      end
+
+      def make_tr_attr_style(cell, row, rowmax, totalrows, opt)
+        top = row.zero? ? "#{SW1} 1.5pt;" : "none;"
+        bottom = "#{SW1} #{rowmax >= totalrows ? '1.5' : '1.0'}pt;"
+        ret = <<~STYLE.gsub(/\n/, "")
+          border-top:#{top}mso-border-top-alt:#{top}
+          border-left:#{bottom}mso-border-top-alt:#{bottom}
+          border-right:#{bottom}mso-border-top-alt:#{bottom}
+          border-bottom:#{bottom}mso-border-bottom-alt:#{bottom}
+        STYLE
+        opt[:bordered] or ret = ""
+        pb = keep_rows_together(cell, rowmax, totalrows, opt) ? "avoid" : "auto"
+        "#{ret}page-break-after:#{pb};"
+      end
+
+      def new_fullcolspan_row(table, tfoot)
+        # how many columns in the table?
+        cols = 0
+        table.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;" \
+                "border-left:#{SW1} 1.5pt;mso-border-left-alt:#{SW1} 1.5pt;" \
+                "border-right:#{SW1} 1.5pt;mso-border-right-alt:#{SW1} 1.5pt;"
+        tfoot.add_child("<tr><td colspan='#{cols}' style='#{style}'/></tr>")
+        tfoot.xpath(".//td").last
+      end
+    end
+  end
+end

data/lib/isodoc/jis/word_cleanup.rb

@@ -0,0 +1,143 @@
+require_relative "../../html2doc/lists"
+
+module IsoDoc
+  module JIS
+    class WordConvert < IsoDoc::Iso::WordConvert
+      def postprocess(result, filename, dir)
+        filename = filename.sub(/\.doc$/, "")
+        header = generate_header(filename, dir)
+        result = from_xhtml(cleanup(to_xhtml(textcleanup(result))))
+        toWord(result, filename, dir, header)
+        @files_to_delete.each { |f| FileUtils.rm_f f }
+      end
+
+      def word_cleanup(docxml)
+        word_note_cleanup(docxml)
+        boldface(docxml)
+        super
+        move_to_inner_cover(docxml)
+      end
+
+      def move_to_inner_cover(docxml)
+        source = docxml.at("//div[@type = 'inner-cover-note']")
+        dest = docxml.at("//div[@id = 'boilerplate-inner-cover-note']")
+        source && dest and dest.replace(source.remove)
+        source = docxml.at("//div[@type = 'contributors']")
+        dest = docxml.at("//div[@id = 'boilerplate-contributors']")
+        source && dest and dest.replace(source.remove)
+        docxml
+      end
+
+      def word_intro(docxml, level)
+        intro = insert_toc(File.read(@wordintropage, encoding: "UTF-8"),
+                           docxml, level)
+        intro = populate_template(intro, :word)
+        introxml = to_word_xhtml_fragment(intro)
+        docxml.at('//div[@class="WordSection2"]') << introxml
+          .to_xml(encoding: "US-ASCII")
+      end
+
+      def word_note_cleanup(docxml)
+        docxml.xpath("//p[@class = 'Note']").each do |p|
+          p.xpath("//following-sibling::p").each do |p2|
+            p2["class"] == "Note" and
+              p2["class"] = "NoteCont"
+          end
+        end
+      end
+
+      def boldface(docxml)
+        docxml.xpath("//b").each do |b|
+          b.name = "span"
+          b["class"] = "Strong"
+        end
+      end
+
+      def toWord(result, filename, dir, header)
+        result = word_split(word_cleanup(to_xhtml(result)))
+        @wordstylesheet = wordstylesheet_update
+        result.each do |k, v|
+          to_word1(v, "#{filename}#{k}", dir, header)
+        end
+        header&.unlink
+        @wordstylesheet.unlink if @wordstylesheet.is_a?(Tempfile)
+      end
+
+      def to_word1(result, filename, dir, header)
+        result or return
+        result = from_xhtml(result).gsub(/-DOUBLE_HYPHEN_ESCAPE-/, "--")
+        ::Html2Doc::JIS.new(
+          filename: filename, imagedir: @localdir,
+          stylesheet: @wordstylesheet&.path,
+          header_file: header&.path, dir: dir,
+          asciimathdelims: [@openmathdelim, @closemathdelim],
+          liststyles: { ul: @ulstyle, ol: @olstyle }
+        ).process(result)
+      end
+
+      def word_split(xml)
+        b = xml.dup
+        { _cover: cover_split(xml), "": main_split(b) }
+      end
+
+      def cover_split(xml)
+        xml.at("//body").elements.each do |e|
+          e.name == "div" && e["class"] == "WordSection1" and next
+          e.remove
+        end
+        xml
+      end
+
+      def main_split(xml)
+        c = xml.at("//div[@class = 'WordSection1']")
+        c.next_element&.remove
+        c.remove
+        c = xml.at("//div[@class = 'WordSection2']")
+        c.elements.first.at("./br") and c.elements.first.remove
+        xml
+      end
+
+      STYLESMAP = {}.freeze
+
+      def style_cleanup(docxml)
+        new_styles(docxml)
+        index_cleanup(docxml)
+      end
+
+      def new_styles(docxml)
+        super
+        biblio_paras(docxml)
+        heading_to_para(docxml)
+      end
+
+      def biblio_paras(docxml)
+        docxml.xpath("//div[@class = 'normref_div']//" \
+                     "p[not(@class) or @class = 'MsoNormal']").each do |p|
+          p["class"] = "NormRefText"
+        end
+      end
+
+      def heading_to_para(docxml)
+        docxml.xpath("//h1[@class = 'ForewordTitle']").each do |p|
+          p.name = "p"
+          p.parent.xpath("./p[not(@class) or @class = 'MsoNormal']").each do |n|
+            n["class"] = "ForewordText"
+          end
+        end
+        docxml.xpath("//h1[@class = 'IntroTitle'] | //h1[@class = 'Annex'] | " \
+                     "//h2[@class = 'Terms'] | " \
+                     "//h3[@class = 'Terms'] | //h4[@class = 'Terms'] | " \
+                     "//h5[@class = 'Terms'] | //h6[@class = 'Terms']").each do |n|
+          n.name = "p"
+        end
+      end
+
+      def word_annex_cleanup1(docxml, lvl)
+        docxml.xpath("//h#{lvl}[ancestor::*[@class = 'Section3']]").each do |h2|
+          h2.name = "p"
+          h2["class"] = "h#{lvl}Annex"
+        end
+      end
+    end
+  end
+end

data/lib/isodoc/jis/word_convert.rb

@@ -1,7 +1,9 @@
-require_relative "../../html2doc/lists"
 require_relative "base_convert"
 require "isodoc"
 require_relative "init"
+require_relative "word_cleanup"
+require_relative "figure"
+require_relative "table"
 
 module IsoDoc
   module JIS
@@ -14,6 +16,11 @@ module IsoDoc
 
       def init_dis(opt); end
 
+      def clause_attrs(node)
+        # capture the type of clause
+        { id: node["id"], type: node["type"] }
+      end
+
       def convert(input_filename, file = nil, debug = false,
                 output_filename = nil)
         file = File.read(input_filename, encoding: "utf-8") if file.nil?
@@ -56,85 +63,10 @@ module IsoDoc
           olstyle: "l8" }
       end
 
-      def postprocess(result, filename, dir)
-        filename = filename.sub(/\.doc$/, "")
-        header = generate_header(filename, dir)
-        result = from_xhtml(cleanup(to_xhtml(textcleanup(result))))
-        toWord(result, filename, dir, header)
-        @files_to_delete.each { |f| FileUtils.rm_f f }
-      end
-
-      def word_cleanup(docxml)
-        word_note_cleanup(docxml)
-        boldface(docxml)
-        super
-      end
-
-      def word_note_cleanup(docxml)
-        docxml.xpath("//p[@class = 'Note']").each do |p|
-          p.xpath("//following-sibling::p").each do |p2|
-            p2["class"] == "Note" and
-              p2["class"] = "NoteCont"
-          end
-        end
-      end
-
-      def boldface(docxml)
-        docxml.xpath("//h1 | h2 | h3 | h4 | h5 | h6").each do |h|
-          h.children = "<b>#{to_xml(h.children)}</b>"
-        end
-        docxml.xpath("//b").each do |b|
-          b.name = "span"
-          b["class"] = "Strong"
-        end
-      end
-
-      def toWord(result, filename, dir, header)
-        result = word_split(word_cleanup(to_xhtml(result)))
-        @wordstylesheet = wordstylesheet_update
-        result.each do |k, v|
-          to_word1(v, "#{filename}#{k}", dir, header)
-        end
-        header&.unlink
-        @wordstylesheet.unlink if @wordstylesheet.is_a?(Tempfile)
-      end
-
-      def to_word1(result, filename, dir, header)
-        result or return
-        result = from_xhtml(result).gsub(/-DOUBLE_HYPHEN_ESCAPE-/, "--")
-        ::Html2Doc::JIS.new(
-          filename: filename, imagedir: @localdir,
-          stylesheet: @wordstylesheet&.path,
-          header_file: header&.path, dir: dir,
-          asciimathdelims: [@openmathdelim, @closemathdelim],
-          liststyles: { ul: @ulstyle, ol: @olstyle }
-        ).process(result)
-      end
-
-      def word_split(xml)
-        b = xml.dup
-        { _cover: cover_split(xml), "": main_split(b) }
-      end
-
-      def cover_split(xml)
-        xml.at("//body").elements.each do |e|
-          e.name == "div" && e["class"] == "WordSection1" and next
-          e.remove
-        end
-        xml
-      end
-
-      def main_split(xml)
-        c = xml.at("//div[@class = 'WordSection1']")
-        c.next_element&.remove
-        c.remove
-        xml
-      end
-
       def norm_ref(isoxml, out, num)
         (f = isoxml.at(ns(norm_ref_xpath)) and f["hidden"] != "true") or
           return num
-        out.div class: "normref" do |div|
+        out.div class: "normref_div" do |div|
           num += 1
           clause_name(f, f.at(ns("./title")), div, nil)
           if f.name == "clause"
@@ -167,24 +99,104 @@ module IsoDoc
         end
       end
 
-      def new_styles(docxml)
+      def preface(isoxml, out)
+        isoxml.xpath(ns("//preface/clause | //preface/references | " \
+                        "//preface/definitions | //preface/terms")).each do |f|
+          out.div **attr_code(class: "Section3", id: f["id"],
+                              type: f["type"]) do |div|
+            clause_name(f, f&.at(ns("./title")), div, { class: "IntroTitle" })
+            f.elements.each do |e|
+              parse(e, div) unless e.name == "title"
+            end
+          end
+        end
+      end
+
+      def introduction(isoxml, out)
+        f = isoxml.at(ns("//introduction")) || return
+        out.div class: "Section3", id: f["id"] do |div|
+          clause_name(f, f.at(ns("./title")), div, { class: "IntroTitle" })
+          f.elements.each do |e|
+            parse(e, div) unless e.name == "title"
+          end
+        end
+      end
+
+      def make_body2(body, docxml)
+        body.div class: "WordSection2" do |div2|
+          boilerplate docxml, div2
+          preface_block docxml, div2
+          abstract docxml, div2
+          foreword docxml, div2
+          preface docxml, div2
+          acknowledgements docxml, div2
+          div2.p { |p| p << "&#xa0;" } # placeholder
+        end
+        section_break(body)
+      end
+
+      def middle(isoxml, out)
+        middle_title(isoxml, out)
+        middle_admonitions(isoxml, out)
+        introduction isoxml, out
+        scope isoxml, out, 0
+        norm_ref isoxml, out, 0
+        clause_etc isoxml, out, 0
+        annex isoxml, out
+        bibliography isoxml, out
+      end
+
+      def make_body3(body, docxml)
         super
-        biblio_paras(docxml)
-        heading_to_para(docxml)
+        commentary docxml, body
+      end
+
+      def footnote_parse(node, out)
+        return table_footnote_parse(node, out) if @in_table || @in_figure # &&
+
+        # !node.ancestors.map(&:name).include?("name")
+
+        fn = node["reference"] || UUIDTools::UUID.random_create.to_s
+        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|
+            a.sup { |sup| sup << fn }
+          end
+        end
+        @in_footnote = true
+        @footnotes << make_generic_footnote_text(node, fn)
+        @in_footnote = false
+        @seen_footnote << fn
+      end
+
+      def annex(isoxml, out)
+        amd(isoxml) and @suppressheadingnumbers = @oldsuppressheadingnumbers
+        isoxml.xpath(ns("//annex[not(@commentary = 'true')]")).each do |c|
+          page_break(out)
+          render_annex(out, c)
+        end
+        amd(isoxml) and @suppressheadingnumbers = true
       end
 
-      def biblio_paras(docxml)
-        docxml.xpath("//div[@class = 'normref']//" \
-                     "p[not(@class) or @class = 'MsoNormal']").each do |p|
-          p["class"] = "NormRefText"
+      def commentary(isoxml, out)
+        isoxml.xpath(ns("//annex[@commentary = 'true']")).each do |c|
+          section_break(out)
+          out.div class: "WordSectionCommentary" do |div|
+            commentary_title(isoxml, div)
+            render_annex(div, c)
+          end
         end
       end
 
-      def heading_to_para(docxml)
-        docxml.xpath("//h1[@class = 'ForewordTitle']").each do |p|
-          p.name = "p"
-          p.xpath("../div/p[not(@class) or @class = 'MsoNormal']").each do |n|
-            n["class"] = "ForewordText"
+      def render_annex(out, clause)
+        out.div **attr_code(annex_attrs(clause)) do |s|
+          clause.elements.each do |c1|
+            if c1.name == "title" then annex_name(clause, c1, s)
+            else parse(c1, s)
+            end
           end
         end
       end

data/lib/isodoc/jis/xref.rb

@@ -1,6 +1,58 @@
 module IsoDoc
   module JIS
+    class Counter < IsoDoc::XrefGen::Counter
+    end
+
     class Xref < IsoDoc::Iso::Xref
+      def annex_name_lbl(clause, num)
+        obl = l10n("(#{@labels['inform_annex']})")
+        clause["obligation"] == "normative" and
+          obl = l10n("(#{@labels['norm_annex']})")
+        title = Common::case_with_markup(@labels["annex"], "capital", @script)
+        l10n("#{title} #{num}<br/>#{obl}")
+      end
+
+      def back_anchor_names(xml)
+        if @parse_settings.empty? || @parse_settings[:clauses]
+          i = Counter.new("@")
+          xml.xpath(ns("//annex")).each do |c|
+            if c["commentary"] == "true"
+              commentary_names(c)
+            else
+              annex_names(c, i.increment(c).print)
+            end
+          end
+          xml.xpath(ns(@klass.bibliography_xpath)).each do |b|
+            preface_names(b)
+          end
+          xml.xpath(ns("//colophon/clause")).each { |b| preface_names(b) }
+          xml.xpath(ns("//indexsect")).each { |b| preface_names(b) }
+        end
+        references(xml) if @parse_settings.empty? || @parse_settings[:refs]
+      end
+
+      def commentary_names(clause)
+        preface_name_anchors(clause, 1, clause_title(clause))
+        clause.xpath(ns(SUBCLAUSES)).each_with_object(Counter.new) do |c, i|
+          commentary_names1(c, clause["id"], i.increment(c).print, 2)
+        end
+      end
+
+      def commentary_names1(clause, root, num, level)
+        commentary_name_anchors(clause, num, root, level)
+        clause.xpath(ns(SUBCLAUSES)).each_with_object(Counter.new) do |c, i|
+          commentary_names1(c, root, "#{num}.#{i.increment(c).print}",
+                            level + 1)
+        end
+      end
+
+      def commentary_name_anchors(clause, num, root, level)
+        @anchors[clause["id"]] =
+          { label: num, xref: l10n("#{@labels['clause']} #{num}"),
+            container: root,
+            title: clause_title(clause), level: level, type: "clause",
+            elem: @labels["clause"] }
+      end
     end
   end
 end