metanorma-gb 1.0.0

data/lib/isodoc/gb/gbhtmlrender.rb

@@ -0,0 +1,208 @@
+module IsoDoc
+  module Gb
+    # A {Converter} implementation that generates GB output, and a document
+    # schema encapsulation of the document for validation
+    class HtmlConvert < IsoDoc::HtmlConvert
+      def formula_parse(node, out)
+        out.div **attr_code(id: node["id"], class: "formula") do |div|
+          insert_tab(div, 1)
+          parse(node.at(ns("./stem")), out)
+          insert_tab(div, 1)
+          div << "(#{get_anchors[node['id']][:label]})"
+        end
+        formula_where(node.at(ns("./dl")), out)
+      end
+
+      def formula_where(dl, out)
+        return unless dl
+        out.p { |p| p << @where_lbl }
+        formula_dl_parse(dl, out)
+      end
+
+      def formula_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" }) { |td| td << "&mdash;" }
+              tr.td **{ valign: "top" } do |listitem|
+                dd.children.each { |n| parse(n, listitem) }
+              end
+            end
+          end
+        end
+      end
+
+      EXAMPLE_TBL_ATTR =
+        { valign: "top", class: "example_label",
+          style: "padding:2pt 2pt 2pt 2pt" }.freeze
+
+      def example_parse(node, out)
+        out.table **attr_code(id: node["id"], class: "example") do |t|
+          t.tr do |tr|
+            tr.td **EXAMPLE_TBL_ATTR do |td|
+              td << l10n(example_label(node) + ":")
+            end
+            tr.td **{ valign: "top", class: "example" } do |td|
+              node.children.each { |n| parse(n, td) }
+            end
+          end
+        end
+      end
+
+      def note_parse(node, out)
+        @note = true
+        out.table **attr_code(id: node["id"], class: "Note") do |t|
+          t.tr do |tr|
+            tr.td **EXAMPLE_TBL_ATTR do |td|
+              td << l10n(note_label(node) + ":")
+            end
+            tr.td **{ valign: "top", class: "Note" } do |td|
+              node.children.each { |n| parse(n, td) }
+            end
+          end
+        end
+        @note = false
+      end
+
+      def termnote_parse(node, out)
+        @note = true
+        out.table **attr_code(id: node["id"], class: "Note") do |t|
+          t.tr do |tr|
+            tr.td **EXAMPLE_TBL_ATTR do |td|
+              td << l10n("#{get_anchors[node['id']][:label]}:")
+            end
+            tr.td **{ valign: "top", class: "Note" } do |td|
+              node.children.each { |n| parse(n, td) }
+            end
+          end
+        end
+        @note = false
+      end
+
+      def middle(isoxml, out)
+        super
+        end_line(isoxml, out)
+      end
+
+      def end_line(_isoxml, out)
+        out.hr **{ width: "25%" }
+      end
+
+      def error_parse(node, out)
+        # catch elements not defined in ISO
+        case node.name
+        when "string" then string_parse(node, out)
+        else
+          super
+        end
+      end
+
+      def string_parse(node, out)
+        if node["script"] == "Hant"
+          out.span **{ class: "Hant" } do |s|
+            node.children.each { |c| parse(c, s) }
+          end
+        else
+          node.children.each { |c| parse(c, out) }
+        end
+      end
+
+      def deprecated_term_parse(node, out)
+        out.p **{ class: "DeprecatedTerms" } do |p|
+          p << l10n("#{@deprecated_lbl}: ")
+          node.children.each { |c| parse(c, p) }
+        end
+      end
+
+      def termref_render(x)
+        parts = x.split(%r{(\s*\[MODIFICATION\]|,)}m)
+        parts[1] = l10n(", #{@source_lbl}") if parts.size > 1 && parts[1] == ","
+        parts.map do |p|
+          /\s*\[MODIFICATION\]/.match?(p) ? l10n(", #{@modified_lbl} &mdash; ") : p
+        end.join.sub(/\A\s*/m, l10n("[")).sub(/\s*\z/m, l10n("]"))
+      end
+
+      def termref_resolve(docxml)
+        docxml.split(%r{(\[TERMREF\]|\[/TERMREF\])}).each_slice(4).
+          map do |a|
+          a.size < 3 ? a[0] : a[0] + termref_render(a[2])
+        end.join
+      end
+
+      def populate_template(docxml, format)
+        meta = @meta.get.merge(@labels)
+        logo = @common.format_logo(meta[:gbprefix], meta[:gbscope], format)
+        logofile = @meta.standard_logo(meta[:gbprefix])
+        @files_to_delete << logofile + ".gif" unless logofile.nil?
+        docxml = termref_resolve(docxml)
+        meta[:standard_agency_formatted] =
+          @common.format_agency(meta[:standard_agency], format)
+        meta[:standard_logo] = logo
+        template = Liquid::Template.parse(docxml)
+        template.render(meta.map { |k, v| [k.to_s, v] }.to_h)
+      end
+
+      def foreword(isoxml, out)
+        f = isoxml.at(ns("//foreword")) || return
+        page_break(out)
+        out.div do |s|
+          s.h1 **{ class: "ForewordTitle" } do |h1|
+            h1 << "#{@foreword_lbl}&nbsp;"
+            # insert_tab(h1, 1)
+          end
+          f.elements.each { |e| parse(e, s) unless e.name == "title" }
+        end
+      end
+
+      def clause_name(num, title, div, header_class)
+        header_class = {} if header_class.nil?
+        div.h1 **attr_code(header_class) do |h1|
+          if num
+            h1 << "#{num}."
+            h1 << "&#x3000;"
+          end
+          h1 << title
+        end
+        div.parent.at(".//h1")
+      end
+
+      def clause_parse_title(node, div, c1, out)
+        if node["inline-header"] == "true"
+          inline_header_title(out, node, c1)
+        else
+          div.send "h#{get_anchors[node['id']][:level]}" do |h|
+            h << "#{get_anchors[node['id']][:label]}.&#x3000;"
+            c1.children.each { |c2| parse(c2, h) }
+          end
+        end
+      end
+
+      def annex_name(annex, name, div)
+        div.h1 **{ class: "Annex" } do |t|
+          t << "#{get_anchors[annex['id']][:label]}<br/><br/>"
+          t << name.text
+        end
+      end
+
+      def term_defs_boilerplate(div, source, term, preface)
+        unless preface
+          (source.empty? && term.nil?) and div << @no_terms_boilerplate or
+            div << term_defs_boilerplate_cont(source, term)
+        end
+        #div << @term_def_boilerplate unless preface
+      end
+
+      def reference_names(ref)
+        isopub = ref.at(ns(ISO_PUBLISHER_XPATH))
+        docid = ref.at(ns("./docidentifier"))
+        date = ref.at(ns("./date[@type = 'published']"))
+        reference = format_ref(docid.text, isopub, date)
+        @anchors[ref["id"]] = { xref: reference }
+      end
+    end
+  end
+end
+

data/lib/isodoc/gb/gbwordconvert.rb

@@ -0,0 +1,136 @@
+require "isodoc"
+require_relative "gbconvert"
+require_relative "gbcleanup"
+require "gb_agencies"
+require_relative "metadata"
+require_relative "gbwordrender"
+require "fileutils"
+
+module IsoDoc
+  module Gb
+    # A {Converter} implementation that generates GB output, and a document
+    # schema encapsulation of the document for validation
+    class WordConvert < IsoDoc::WordConvert
+      def initialize(options)
+        @common = IsoDoc::Gb::Common.new(options)
+        @libdir = File.dirname(__FILE__)
+        super
+        @lang = "zh"
+        @script = "Hans"
+      end
+        
+      def default_fonts(options)
+        script = options[:script] || "Hans"
+        scope = options[:scope] || "national"
+        {
+          bodyfont: (script == "Hans" ? '"SimSun",serif' : '"Cambria",serif'),
+          headerfont: (script == "Hans" ? '"SimHei",sans-serif' : '"Calibri",sans-serif'),
+          monospacefont: '"Courier New",monospace',
+          titlefont: (scope == "national" ? (script != "Hans" ? '"Cambria",serif' : '"SimSun",serif' ) :
+           (script == "Hans" ? '"SimHei",sans-serif' : '"Calibri",sans-serif' ))
+        }     
+      end     
+                
+      def default_file_locations(options)
+        {   
+          wordstylesheet: html_doc_path("wordstyle.scss"),
+          standardstylesheet: html_doc_path("gb.scss"),
+          header: html_doc_path("header.html"),
+          wordcoverpage: html_doc_path("word_gb_titlepage.html"),
+          wordintropage: html_doc_path("word_gb_intro.html"),
+          ulstyle: "l7",
+          olstyle: "l10",
+        }
+      end 
+          
+      def extract_fonts(options)
+        b = options[:bodyfont] || "Arial"
+        h = options[:headerfont] || "Arial"
+        m = options[:monospacefont] || "Courier"
+        t = options[:titlefont] || "Arial"
+        "$bodyfont: #{b};\n$headerfont: #{h};\n$monospacefont: #{m};\n$titlefont: #{t};\n"
+      end   
+
+      def metadata_init(lang, script, labels)
+        unless ["en", "zh"].include? lang
+          lang = "zh"
+          script = "Hans"
+        end
+        @meta = Metadata.new(lang, script, labels)
+        @common = IsoDoc::Gb::Common.new(meta: @meta)
+      end
+
+      def cleanup(docxml)
+        @cleanup = Cleanup.new(@script, @deprecated_lbl)
+        super
+        @cleanup.cleanup(docxml)
+        docxml
+      end
+
+      def example_cleanup(docxml)
+        super
+        @cleanup.example_cleanup(docxml)
+      end
+
+      def i18n_init(lang, script)
+        super
+        y = if lang == "en"
+              YAML.load_file(File.join(File.dirname(__FILE__), "i18n-en.yaml"))
+            elsif lang == "zh" && script == "Hans"
+              YAML.load_file(File.join(File.dirname(__FILE__),
+                                       "i18n-zh-Hans.yaml"))
+            else
+              YAML.load_file(File.join(File.dirname(__FILE__), "i18n-zh-Hans.yaml"))
+            end
+        @labels = @labels.merge(y)
+      end
+
+      ENDLINE = <<~END.freeze
+      <v:line id="_x0000_s1026"
+ alt="" style='position:absolute;left:0;text-align:left;z-index:251662848;
+ mso-wrap-edited:f;mso-width-percent:0;mso-height-percent:0;
+ mso-width-percent:0;mso-height-percent:0'
+ from="6.375cm,20.95pt" to="10.625cm,20.95pt"
+ strokeweight="1.5pt"/>
+      END
+
+      def end_line(_isoxml, out)
+        out.parent.add_child(ENDLINE)
+      end
+
+      def generate_header(filename, dir)
+        return unless @header
+        template = Liquid::Template.parse(File.read(@header, encoding: "UTF-8"))
+        meta = @meta.get
+        meta[:filename] = filename
+        params = meta.map { |k, v| [k.to_s, v] }.to_h
+        File.open("header.html", "w") { |f| f.write(template.render(params)) }
+        system "cp #{@common.fileloc(File.join('html', 'blank.png'))} blank.png"
+        FileUtils.cp @common.fileloc(File.join('html', 'blank.png')), "blank.png"
+        @files_to_delete << "blank.png"
+        @files_to_delete << "header.html"
+        "header.html"
+      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.replace(" ") if x.name == "span" &&
+            /mso-tab-count/.match?(x["style"])
+          x.remove if x.name == "span" && x["class"] == "MsoCommentReference"
+          x.remove if x.name == "a" && x["epub:type"] == "footnote"
+          x.replace(x.children) if x.name == "a"
+        end
+        from_xhtml(h1)
+      end
+
+      def word_cleanup(docxml)
+        word_preface(docxml)
+        word_annex_cleanup(docxml)
+        @cleanup.title_cleanup(docxml.at('//div[@class="WordSection2"]'))
+        docxml
+      end
+    end
+  end
+end

data/lib/isodoc/gb/gbwordrender.rb

@@ -0,0 +1,206 @@
+module IsoDoc
+  module Gb
+    # A {Converter} implementation that generates GB output, and a document
+    # schema encapsulation of the document for validation
+    class WordConvert < IsoDoc::WordConvert
+      def formula_parse(node, out)
+        out.div **attr_code(id: node["id"], class: "formula") do |div|
+          insert_tab(div, 1)
+          parse(node.at(ns("./stem")), out)
+          insert_tab(div, 1)
+          div << "(#{get_anchors[node['id']][:label]})"
+        end
+        formula_where(node.at(ns("./dl")), out)
+      end
+
+      def formula_where(dl, out)
+        return unless dl
+        out.p { |p| p << @where_lbl }
+        formula_dl_parse(dl, out)
+      end
+
+      def formula_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" }) { |td| td << "&mdash;" }
+              tr.td **{ valign: "top" } do |listitem|
+                dd.children.each { |n| parse(n, listitem) }
+              end
+            end
+          end
+        end
+      end
+
+      EXAMPLE_TBL_ATTR =
+        { valign: "top", class: "example_label",
+          style: "padding:2pt 2pt 2pt 2pt" }.freeze
+
+      def example_parse(node, out)
+        out.table **attr_code(id: node["id"], class: "example") do |t|
+          t.tr do |tr|
+            tr.td **EXAMPLE_TBL_ATTR do |td|
+              td << l10n(example_label(node) + ":")
+            end
+            tr.td **{ valign: "top", class: "example" } do |td|
+              node.children.each { |n| parse(n, td) }
+            end
+          end
+        end
+      end
+
+      def note_parse(node, out)
+        @note = true
+        out.table **attr_code(id: node["id"], class: "Note") do |t|
+          t.tr do |tr|
+            tr.td **EXAMPLE_TBL_ATTR do |td|
+              td << l10n(note_label(node) + ":")
+            end
+            tr.td **{ valign: "top", class: "Note" } do |td|
+              node.children.each { |n| parse(n, td) }
+            end
+          end
+        end
+        @note = false
+      end
+
+      def termnote_parse(node, out)
+        @note = true
+        out.table **attr_code(id: node["id"], class: "Note") do |t|
+          t.tr do |tr|
+            tr.td **EXAMPLE_TBL_ATTR do |td|
+              td << l10n("#{get_anchors[node['id']][:label]}:")
+            end
+            tr.td **{ valign: "top", class: "Note" } do |td|
+              node.children.each { |n| parse(n, td) }
+            end
+          end
+        end
+        @note = false
+      end
+
+      def middle(isoxml, out)
+        super
+        end_line(isoxml, out)
+      end
+
+      def end_line(_isoxml, out)
+        out.hr **{ width: "25%" }
+      end 
+
+      def error_parse(node, out)
+        # catch elements not defined in ISO
+        case node.name
+        when "string" then string_parse(node, out)
+        else
+          super
+        end
+      end
+
+      def string_parse(node, out)
+        if node["script"] == "Hant"
+          out.span **{ class: "Hant" } do |s|
+            node.children.each { |c| parse(c, s) }
+          end
+        else
+          node.children.each { |c| parse(c, out) }
+        end
+      end
+
+      def deprecated_term_parse(node, out) 
+        out.p **{ class: "DeprecatedTerms" } do |p|
+          p << l10n("#{@deprecated_lbl}: ")
+          node.children.each { |c| parse(c, p) }
+        end
+      end
+
+      def termref_render(x)
+        parts = x.split(%r{(\s*\[MODIFICATION\]|,)}m)
+        parts[1] = l10n(", #{@source_lbl}") if parts.size > 1 && parts[1] == ","
+        parts.map do |p|
+          /\s*\[MODIFICATION\]/.match?(p) ? l10n(", #{@modified_lbl} &mdash; ") : p
+        end.join.sub(/\A\s*/m, l10n("[")).sub(/\s*\z/m, l10n("]"))
+      end
+
+      def termref_resolve(docxml)
+        docxml.split(%r{(\[TERMREF\]|\[/TERMREF\])}).each_slice(4).
+          map do |a|
+          a.size < 3 ? a[0] : a[0] + termref_render(a[2])
+        end.join
+      end
+
+      def populate_template(docxml, format)
+        meta = @meta.get.merge(@labels)
+        logo = @common.format_logo(meta[:gbprefix], meta[:gbscope], format)
+        logofile = @meta.standard_logo(meta[:gbprefix])
+        @files_to_delete << logofile + ".gif" unless logofile.nil?
+        docxml = termref_resolve(docxml)
+        meta[:standard_agency_formatted] =
+          @common.format_agency(meta[:standard_agency], format)
+        meta[:standard_logo] = logo
+        template = Liquid::Template.parse(docxml)
+        template.render(meta.map { |k, v| [k.to_s, v] }.to_h)
+      end
+
+      def foreword(isoxml, out)
+        f = isoxml.at(ns("//foreword")) || return
+        page_break(out)
+        out.div do |s|
+          s.h1 **{ class: "ForewordTitle" } do |h1|
+            h1 << "#{@foreword_lbl}&nbsp;"
+            # insert_tab(h1, 1)
+          end
+          f.elements.each { |e| parse(e, s) unless e.name == "title" }
+        end
+      end
+
+      def clause_name(num, title, div, header_class)
+        header_class = {} if header_class.nil?
+        div.h1 **attr_code(header_class) do |h1|
+          if num
+            h1 << "#{num}."
+            h1 << "&#x3000;"
+          end
+          h1 << title
+        end
+        div.parent.at(".//h1")
+      end
+
+      def clause_parse_title(node, div, c1, out)
+        if node["inline-header"] == "true"
+          inline_header_title(out, node, c1)
+        else
+          div.send "h#{get_anchors[node['id']][:level]}" do |h|
+            h << "#{get_anchors[node['id']][:label]}.&#x3000;"
+            c1.children.each { |c2| parse(c2, h) }
+          end
+        end
+      end
+
+      def annex_name(annex, name, div)
+        div.h1 **{ class: "Annex" } do |t|
+          t << "#{get_anchors[annex['id']][:label]}<br/><br/>"
+          t << name.text
+        end
+      end
+
+      def term_defs_boilerplate(div, source, term, preface)
+        (source.empty? && term.nil?) and div << @no_terms_boilerplate or
+          div << term_defs_boilerplate_cont(source, term)
+        #div << @term_def_boilerplate unless preface
+      end
+
+      def reference_names(ref)
+        isopub = ref.at(ns(ISO_PUBLISHER_XPATH))
+        docid = ref.at(ns("./docidentifier"))
+        date = ref.at(ns("./date[@type = 'published']"))
+        reference = format_ref(docid.text, isopub, date)
+        @anchors[ref["id"]] = { xref: reference }
+      end
+    end
+  end
+end
+