metanorma-iso 2.0.3 → 2.0.6

data/lib/metanorma/iso/validate_list.rb

@@ -0,0 +1,122 @@
+module Metanorma
+  module ISO
+    class Converter < Standoc::Converter
+      # https://www.iso.org/ISO-house-style.html#iso-hs-s-text-r-p-lists
+      def listcount_validate(doc)
+        return if @novalid
+
+        ol_count_validate(doc)
+        li_depth_validate(doc)
+      end
+
+      def ol_count_validate(doc)
+        doc.xpath("//clause | //annex").each do |c|
+          next if c.xpath(".//ol").empty?
+
+          ols = c.xpath(".//ol") -
+            c.xpath(".//ul//ol | .//ol//ol | .//clause//ol")
+          ols.size > 1 and
+            style_warning(c, "More than 1 ordered list in a numbered clause",
+                          nil)
+        end
+      end
+
+      def li_depth_validate(doc)
+        doc.xpath("//li//li//li//li").each do |l|
+          l.at(".//li") and
+            style_warning(l, "List more than four levels deep", nil)
+        end
+      end
+
+      # https://www.iso.org/ISO-house-style.html#iso-hs-s-text-r-p-lists
+      def list_punctuation(doc)
+        return if @novalid
+
+        ((doc.xpath("//ol") - doc.xpath("//ul//ol | //ol//ol")) +
+         (doc.xpath("//ul") - doc.xpath("//ul//ul | //ol//ul"))).each do |list|
+          next if skip_list_punctuation(list)
+
+          prec = list.at("./preceding::text()[normalize-space(.) != ''][1]")
+          list_punctuation1(list, prec&.text)
+        end
+      end
+
+      def skip_list_punctuation(list)
+        return true if list.at("./ancestor::table")
+        return true if list.at("./following-sibling::term") # terms boilerplate
+
+        list.xpath(".//li").each do |entry|
+          l = entry.dup
+          l.xpath(".//ol | .//ul").each(&:remove)
+          l.text.split.size > 2 and return false
+        end
+        true
+      end
+
+      def list_punctuation1(list, prectext)
+        prectext ||= ""
+        entries = list.xpath(".//li")
+        case prectext.strip.chars.last
+        when ":", "" then list_after_colon_punctuation(list, entries)
+        when "." then entries.each { |li| list_full_sentence(li) }
+        else style_warning(list, "All lists must be preceded by "\
+                                 "colon or full stop", prectext)
+        end
+      end
+
+      # if first list entry starts lowercase, treat as sentence broken up
+      def list_after_colon_punctuation(list, entries)
+        lower = starts_lowercase?(list.at(".//li").text)
+        entries.each_with_index do |li, i|
+          if lower
+            list_semicolon_phrase(li, i == entries.size - 1)
+          else
+            list_full_sentence(li)
+          end
+        end
+      end
+
+      def list_semicolon_phrase(elem, last)
+        text = elem.text.strip
+        starts_lowercase?(text) or
+          style_warning(elem, "List entry of broken up sentence must start "\
+                              "with lowercase letter", text)
+        list_semicolon_phrase_punct(elem, text, last)
+      end
+
+      def list_semicolon_phrase_punct(elem, text, last)
+        punct = text.strip.sub(/^.*?(\S)$/m, "\\1")
+        if last
+          punct == "." or
+            style_warning(elem, "Final list entry of broken up "\
+                                "sentence must end with full stop", text)
+        else
+          punct == ";" or
+            style_warning(elem, "List entry of broken up sentence must "\
+                                "end with semicolon", text)
+        end
+      end
+
+      def list_full_sentence(elem)
+        text = elem.text.strip
+        starts_uppercase?(text) or
+          style_warning(elem, "List entry of separate sentences must start "\
+                              "with uppercase letter", text)
+        punct = text.strip.sub(/^.*?(\S)$/m, "\\1")
+        punct == "." or
+          style_warning(elem, "List entry of separate sentences must "\
+                              "end with full stop", text)
+      end
+
+      # allow that all-caps word (acronym) is agnostic as to lowercase
+      def starts_lowercase?(text)
+        text.match?(/^[^[[:upper:]][[:lower:]]]*[[:lower:]]/) ||
+          text.match?(/^[^[[:upper:]][[:lower:]]]*[[:upper:]][[:upper:]]+[^[[:alpha:]]]/)
+      end
+
+      def starts_uppercase?(text)
+        text.match?(/^[^[[:upper:]][[:lower:]]]*[[:upper:]]/)
+      end
+    end
+  end
+end

data/lib/metanorma/iso/validate_section.rb

@@ -14,6 +14,7 @@ module Metanorma
         end
         section_style(doc.root)
         subclause_validate(doc.root)
+        @vocab and vocab_terms_titles_validate(doc.root)
         super
       end
 
@@ -33,7 +34,6 @@ module Metanorma
 
       ONE_SYMBOLS_WARNING = "Only one Symbols and Abbreviated "\
                             "Terms section in the standard".freeze
-
       NON_DL_SYMBOLS_WARNING = "Symbols and Abbreviated Terms can "\
                                "only contain a definition list".freeze
 
@@ -42,11 +42,12 @@ module Metanorma
         f.empty? && return
         (f.size == 1 || @vocab) or
           @log.add("Style", f.first, ONE_SYMBOLS_WARNING)
-        f.first.elements.each do |e|
-          unless %w(title dl).include? e.name
-            @log.add("Style", f.first, NON_DL_SYMBOLS_WARNING)
-            return
-          end
+        f.first.elements.reject { |e| %w(title dl).include? e.name }.empty? or
+          @log.add("Style", f.first, NON_DL_SYMBOLS_WARNING)
+        @vocab and f.each do |f1|
+          f1.at("./ancestor::annex") or
+            @log.add("Style", f1, "In vocabulary documents, Symbols and "\
+                                  "Abbreviated Terms are only permitted in annexes")
         end
       end
 
@@ -72,26 +73,17 @@ module Metanorma
 
       # spec of permissible section sequence
       # we skip normative references, it goes to end of list
-      SEQ =
-        [
-          {
-            msg: "Initial section must be (content) Foreword",
-            val: ["./self::foreword"],
-          },
-          {
-            msg: "Prefatory material must be followed by (clause) Scope",
-            val: ["./self::introduction", "./self::clause[@type = 'scope']"],
-          },
-          {
-            msg: "Prefatory material must be followed by (clause) Scope",
-            val: ["./self::clause[@type = 'scope']"],
-          },
-          {
-            msg: "Normative References must be followed by "\
-                 "Terms and Definitions",
-            val: ["./self::terms | .//terms"],
-          },
-        ].freeze
+      SEQ = [
+        { msg: "Initial section must be (content) Foreword",
+          val: ["./self::foreword"] },
+        { msg: "Prefatory material must be followed by (clause) Scope",
+          val: ["./self::introduction", "./self::clause[@type = 'scope']"] },
+        { msg: "Prefatory material must be followed by (clause) Scope",
+          val: ["./self::clause[@type = 'scope']"] },
+        { msg: "Normative References must be followed by "\
+               "Terms and Definitions",
+          val: ["./self::terms | .//terms"] },
+      ].freeze
 
       SECTIONS_XPATH =
         "//foreword | //introduction | //sections/terms | .//annex | "\
@@ -198,11 +190,6 @@ module Metanorma
         end
       end
 
-      ASSETS_TO_STYLE =
-        "//termsource | //formula | //termnote | "\
-        "//p[not(ancestor::boilerplate)] | //li[not(p)] | //dt | "\
-        "//dd[not(p)] | //td[not(p)] | //th[not(p)]".freeze
-
       NORM_BIBITEMS =
         "//references[@normative = 'true']/bibitem".freeze
 
@@ -217,7 +204,9 @@ module Metanorma
 
       def asset_style(root)
         root.xpath("//example | //termexample").each { |e| example_style(e) }
-        root.xpath("//definition/verbal-definition").each { |e| definition_style(e) }
+        root.xpath("//definition/verbal-definition").each do |e|
+          definition_style(e)
+        end
         root.xpath("//note").each { |e| note_style(e) }
         root.xpath("//fn").each { |e| footnote_style(e) }
         root.xpath(ASSETS_TO_STYLE).each { |e| style(e, extract_text(e)) }
@@ -243,6 +232,24 @@ module Metanorma
           @log.add("Style", nil, "#{location}: subclause is only child")
         end
       end
+
+      # https://www.iso.org/ISO-house-style.html#iso-hs-s-formatting-r-vocabulary
+      def vocab_terms_titles_validate(root)
+        terms = root.xpath("//sections/terms | //sections/clause[.//terms]")
+        if terms.size == 1
+          ((t = terms.first.at("./title")) && (t&.text == @i18n.termsdef)) or
+            @log.add("Style", terms.first,
+                     "Single terms clause in vocabulary document "\
+                     "should have normal Terms and definitions heading")
+        elsif terms.size > 1
+          terms.each do |x|
+            ((t = x.at("./title")) && /^#{@i18n.termsrelated}/.match?(t&.text)) or
+              @log.add("Style", x,
+                       "Multiple terms clauses in vocabulary document "\
+                       "should have 'Terms related to' heading")
+          end
+        end
+      end
     end
   end
 end

data/lib/metanorma/iso/validate_style.rb

@@ -70,8 +70,8 @@ module Metanorma
         style(node, extract_text(node))
       end
 
-      def style_regex(regex, warning, n, text)
-        (m = regex.match(text)) && style_warning(n, warning, m[:num])
+      def style_regex(regex, warning, node, text)
+        (m = regex.match(text)) && style_warning(node, warning, m[:num])
       end
 
       # style check with a regex on a token
@@ -96,10 +96,12 @@ module Metanorma
         style_percent(node, text)
         style_abbrev(node, text)
         style_units(node, text)
+        style_punct(node, text)
       end
 
       # ISO/IEC DIR 2, 9.1
       # ISO/IEC DIR 2, Table B.1
+      # https://www.iso.org/ISO-house-style.html#iso-hs-s-text-r-n-numbers
       def style_number(node, text)
         style_two_regex_not_prev(
           node, text, /^(?<num>-?[0-9]{4,}[,0-9]*)\Z/,
@@ -110,6 +112,8 @@ module Metanorma
                     "possible decimal point", node, text)
         style_regex(/\b(?<num>billions?)\b/i,
                     "ambiguous number", node, text)
+        style_regex(/(^|\s)(?<num>-[0-9][0-9,.]*)/i,
+                    "hyphen instead of minus sign U+2212", node, text)
       end
 
       # ISO/IEC DIR 2, 9.2.1
@@ -158,6 +162,15 @@ module Metanorma
         end
       end
 
+      # https://www.iso.org/ISO-house-style.html#iso-hs-s-text-r-p-and
+      # https://www.iso.org/ISO-house-style.html#iso-hs-s-text-r-p-andor
+      def style_punct(node, text)
+        style_regex(/\b(?<num>and\/?or)\b/i,
+                    "Use 'either x or y, or both'", node, text)
+        style_regex(/\s(?<num>&)\s/i,
+                    "Avoid ampersand in ordinary text'", node, text)
+      end
+
       def style_warning(node, msg, text = nil)
         return if @novalid
 
@@ -165,6 +178,23 @@ module Metanorma
         w += ": #{text}" if text
         @log.add("Style", node, w)
       end
+
+      ASSETS_TO_STYLE =
+        "//termsource | //formula | //termnote | "\
+        "//p[not(ancestor::boilerplate)] | //li[not(p)] | //dt | "\
+        "//dd[not(p)] | //td[not(p)] | //th[not(p)]".freeze
+
+      def asset_style(root)
+        root.xpath("//example | //termexample").each { |e| example_style(e) }
+        root.xpath("//definition/verbal-definition").each do |e|
+          definition_style(e)
+        end
+        root.xpath("//note").each { |e| note_style(e) }
+        root.xpath("//fn").each { |e| footnote_style(e) }
+        root.xpath(ASSETS_TO_STYLE).each { |e| style(e, extract_text(e)) }
+        norm_bibitem_style(root)
+        super
+      end
     end
   end
 end

data/lib/metanorma/iso/validate_title.rb

@@ -68,7 +68,7 @@ module Metanorma
           title = s&.at("./title")&.text || s.name
           s.xpath("./clause | ./terms | ./references").each do |ss|
             subtitle = ss.at("./title")
-            !subtitle.nil? && !subtitle&.text&.empty? or
+            (!subtitle.nil? && !subtitle&.text&.empty?) or
               @log.add("Style", ss,
                        "#{title}: each first-level subclause must have a title")
           end
@@ -91,6 +91,17 @@ module Metanorma
                    "#{label}: all subclauses must have a title, or none")
       end
 
+      # https://www.iso.org/ISO-house-style.html#iso-hs-s-text-r-p-full
+      def title_no_full_stop_validate(root)
+        root.xpath("//preface//title | //sections//title | //annex//title | "\
+                   "//references/title | //preface//name | //sections//name | "\
+                   "//annex//name").each do |t|
+          style_regex(/\A(?<num>.+\.\Z)/i,
+                      "No full stop at end of title or caption",
+                      t, t.text.strip)
+        end
+      end
+
       def title_validate(root)
         title_intro_validate(root)
         title_main_validate(root)
@@ -99,6 +110,7 @@ module Metanorma
         title_names_type_validate(root)
         title_first_level_validate(root)
         title_all_siblings(root.xpath(SECTIONS_XPATH), "(top level)")
+        title_no_full_stop_validate(root)
       end
     end
   end

data/lib/metanorma/iso/version.rb

@@ -1,5 +1,5 @@
 module Metanorma
   module ISO
-    VERSION = "2.0.3".freeze
+    VERSION = "2.0.6".freeze
   end
 end

data/metanorma-iso.gemspec

@@ -46,6 +46,6 @@ Gem::Specification.new do |spec|
   spec.add_development_dependency "sassc", "2.4.0"
   spec.add_development_dependency "simplecov", "~> 0.15"
   spec.add_development_dependency "timecop", "~> 0.9"
-  spec.add_development_dependency "vcr", "~> 5.0.0"
+  spec.add_development_dependency "vcr", "~> 6.1.0"
   spec.add_development_dependency "webmock"
 end

data/spec/isodoc/amd_spec.rb

@@ -74,6 +74,7 @@ RSpec.describe IsoDoc do
           <bibdata>
             <ext>
               <doctype language="">amendment</doctype>
+              <doctype language='en'>Amendment</doctype>
             </ext>
           </bibdata>
           <preface>
@@ -243,9 +244,9 @@ RSpec.describe IsoDoc do
           <bibdata>
             <ext>
               <doctype language=''>amendment</doctype>
+              <doctype language='en'>Amendment</doctype>
             </ext>
           </bibdata>
-          
           <preface>
             <foreword obligation='informative' displayorder='1'>
               <title>Foreword</title>
@@ -436,6 +437,7 @@ RSpec.describe IsoDoc do
         <bibdata>
           <ext>
             <doctype language="">amendment</doctype>
+            <doctype language='en'>Amendment</doctype>
           </ext>
         </bibdata>
         <boilerplate>
@@ -636,10 +638,13 @@ RSpec.describe IsoDoc do
         </body>
       </html>
     OUTPUT
-    expect(xmlpp(IsoDoc::Iso::PresentationXMLConvert.new({}).convert("test", input, true))
-      .sub(%r{<localized-strings>.*</localized-strings>}m, "")).to be_equivalent_to xmlpp(presxml)
-    expect(xmlpp(IsoDoc::Iso::HtmlConvert.new({}).convert("test", presxml, 
-true))).to be_equivalent_to xmlpp(html)
+    expect(xmlpp(IsoDoc::Iso::PresentationXMLConvert.new({})
+      .convert("test", input, true))
+      .sub(%r{<localized-strings>.*</localized-strings>}m, ""))
+      .to be_equivalent_to xmlpp(presxml)
+    expect(xmlpp(IsoDoc::Iso::HtmlConvert.new({})
+      .convert("test", presxml, true)))
+      .to be_equivalent_to xmlpp(html)
   end
 
   it "processes IsoXML metadata" do