require "asciidoctor" require "asciidoctor/iso/converter" require "asciidoctor/gb/version" require "isodoc/gb/gbconvert" require "isodoc/gb/gbwordconvert" require_relative "./section_input.rb" require_relative "./front.rb" require_relative "./validate.rb" require "pp" # require "byebug" module Asciidoctor module Gb GB_NAMESPACE = "http://riboseinc.com/gbstandard" # A {Converter} implementation that generates GB output, and a document # schema encapsulation of the document for validation class Converter < ISO::Converter register_for "gb" def makexml(node) result = ["\n"] @draft = node.attributes.has_key?("draft") @keepboilerplate = node.attributes.has_key?("keep-boilerplate") result << noko { |ixml| front node, ixml } result << noko { |ixml| middle node, ixml } result << "" result = textcleanup(result.flatten * "\n") ret1 = cleanup(Nokogiri::XML(result)) validate(ret1) ret1.root.add_namespace(nil, GB_NAMESPACE) ret1 end def html_doc_path(file) File.join(File.dirname(__FILE__), File.join("html", file)) end def html_converter(node) node.nil? ? IsoDoc::Gb::HtmlConvert.new({}) : IsoDoc::Gb::HtmlConvert.new( script: node.attr("script"), bodyfont: node.attr("body-font"), headerfont: node.attr("header-font"), monospacefont: node.attr("monospace-font"), titlefont: node.attr("title-font"), i18nyaml: node.attr("i18nyaml"), scope: node.attr("scope"), ) end def doc_converter(node) node.nil? ? IsoDoc::Gb::WordConvert.new({}) : IsoDoc::Gb::WordConvert.new( script: node.attr("script"), bodyfont: node.attr("body-font"), headerfont: node.attr("header-font"), monospacefont: node.attr("monospace-font"), titlefont: node.attr("title-font"), i18nyaml: node.attr("i18nyaml"), scope: node.attr("scope"), ) end def default_fonts(node) script = node.attr("script") || "Hans" b = node.attr("body-font") || (script == "Hans" ? '"SimSun",serif' : script == "Latn" ? '"Cambria",serif' : '"SimSun",serif' ) h = node.attr("header-font") || (script == "Hans" ? '"SimHei",sans-serif' : script == "Latn" ? '"Calibri",sans-serif' : '"SimHei",sans-serif' ) m = node.attr("monospace-font") || '"Courier New",monospace' scope = node.attr("scope") || "national" t = node.attr("title-font") || (scope == "national" ? (script != "Hans" ? '"Cambria",serif' : '"SimSun",serif' ) : (script == "Hans" ? '"SimHei",sans-serif' : '"Calibri",sans-serif' )) "$bodyfont: #{b};\n$headerfont: #{h};\n$monospacefont: #{m};\n$titlefont: #{t};\n" end def document(node) init(node) ret = makexml(node).to_xml(indent: 2) unless node.attr("nodoc") || !node.attr("docfile") filename = node.attr("docfile").gsub(/\.adoc$/, "").gsub(%r{^.*/}, "") File.open(filename + ".xml", "w") { |f| f.write(ret) } html_converter(node).convert(filename + ".xml") doc_converter(node).convert(filename + ".xml") end @files_to_delete.each { |f| system "rm #{f}" } ret end def termdef_cleanup(xmldoc) super localisedstr(xmldoc) end ROMAN_TEXT = /\s*[a-z\u00c0-\u00d6\u00d8-\u00f0\u0100-\u0240]/i HAN_TEXT = /\s*[\u4e00-\u9fff]+/ LOCALISED_ELEMS = "//admitted | //deprecates | //preferred | //prefix | "\ "//initial | //addition | //surname | //forename | //name | "\ "//abbreviation | //role/description | //affiliation/description | "\ "//bibdata/item | //bibitem/title | //bibdata/formattedref | "\ "//bibitem/formattedref | //bibdata/note | //bibitem/note | "\ "//bibdata/abstract | //bibitem/note ".freeze MUST_LOCALISE_ELEMS = %w{admitted deprecates preferred}.freeze def localisedstr(xmldoc) xmldoc.xpath(LOCALISED_ELEMS).each do |zh| if zh.at("./string") extract_localisedstrings(zh) elsif MUST_LOCALISE_ELEMS.include? zh.name duplicate_localisedstrings(zh) end end end # element consists solely of localised strings, with no attributes def extract_localisedstrings(elem) elem.xpath("./string").each do |s| s.name = elem.name end elem.replace(elem.children) end def duplicate_localisedstrings(zh) en = zh.dup.remove zh.after(en).after(" ") zh["language"] = "zh" en["language"] = "en" en.traverse do |c| c.text? && c.content = c.text.gsub(HAN_TEXT, "").gsub(/^\s*/, "") end zh.traverse do |c| c.text? && c.content = c.text.gsub(ROMAN_TEXT, "").gsub(/^\s*/, "") end end def inline_quoted(node) ret = noko do |xml| case node.role when "en" then xml.string node.text, **{ language: "en" } when "zh" then xml.string node.text, **{ language: "zh" } when "zh-Hans" xml.string node.text, **{ language: "zh", script: "Hans" } when "zh-Hant" xml.string node.text, **{ language: "zh", script: "Hant" } else nil end end.join return ret unless ret.nil? or ret.empty? super end def termdef_boilerplate_cleanup(xmldoc) return if @keepboilerplate super end GBCODE = "((AQ|BB|CB|CH|CJ|CY|DA|DB|DL|DZ|EJ|FZ|GA|GH|GM|GY|HB|HG|"\ "HJ|HS|HY|JB|JC|JG|JR|JT|JY|LB|LD|LS|LY|MH|MT|MZ|NY|QB|QC|QJ|"\ "QZ|SB|SC|SH|SJ|SN|SY|TB|TD|TJ|TY|WB|WH|WJ|WM|WS|WW|XB|YB|YC|"\ "YD|YS|YY|YZ|ZY|GB|GBZ|GJB|GBn|GHZB|GWKB|GWPB|JJF|JJG|Q|T)(/Z|/T)?)" ISO_REF = %r{^[^"]+)"> \[(?(ISO|IEC|#{GBCODE})[^0-9]*\s[0-9-]+)(:(?[0-9]+))?\],?\s (?.*)$}xm ISO_REF_NO_YEAR = %r{^[^"]+)"> \[(?(ISO|IEC|#{GBCODE})[^0-9]*\s[0-9-]+):--\],?\s? ]*>\s*

(?[^\]]+)

\s*,?\s?(?.*)$}xm ISO_REF_ALL_PARTS = %r{^[^"]+)"> \[(?(ISO|IEC|#{GBCODE})[^0-9]*\s[0-9]+)\s\(all\sparts\)\](

)?,?\s? (?.*)(

)?$}xm def reference1_matches(item) matched = ISO_REF.match item matched2 = ISO_REF_NO_YEAR.match item matched3 = ISO_REF_ALL_PARTS.match item [matched, matched2, matched3] end def cleanup(xmldoc) lang = xmldoc.at("//language")&.text @agencyclass = IsoDoc::Gb::Agencies.new(lang, {}, "") super contributor_cleanup(xmldoc) xmldoc end def docidentifier_cleanup(xmldoc) #byebug id = xmldoc.at("//bibdata/docidentifier/project-number") or return scope = xmldoc.at("//gbscope")&.text prefix = xmldoc.at("//gbprefix")&.text mandate = xmldoc.at("//gbmandate")&.text || "mandatory" idtext = @agencyclass.docidentifier(scope, prefix, mandate, nil, id.text) || return id.content = idtext.gsub(/\ /, " ") end def contributor_cleanup(xmldoc) issuer = xmldoc.at("//bibdata/contributor[role/@type = 'issuer']/organization/name") scope = xmldoc.at("//gbscope")&.text prefix = xmldoc.at("//gbprefix")&.text mandate = xmldoc.at("//gbmandate")&.text || "mandatory" agency = issuer.content agency = @agencyclass.standard_agency1(scope, prefix, mandate) if agency == "GB" agency = "GB" if agency.nil? || agency.empty? owner = xmldoc.at("//copyright/owner/organization/name") owner.content = agency owner = xmldoc.at("//contributor[role/@type = 'issuer']/organization/name") owner.content = agency xmldoc.xpath("//gbcommittee").each do |c| xmldoc.at("//bibdata/contributor").next = ""\ "#{c.text}" end end end end end