lib/govspeak.rb in govspeak-6.7.1 vs lib/govspeak.rb in govspeak-6.7.2

@@ -60,22 +60,34 @@
       @locale = options.fetch(:locale, "en")
       @options = { input: PARSER_CLASS_NAME,
                    sanitize: true,
                    syntax_highlighter: nil }.merge(options)
       @options[:entity_output] = :symbolic
+      @footnote_definition_html = nil
+      @acronyms = []
     end
 
     def to_html
       @to_html ||= begin
-                     html = if @options[:sanitize]
-                              HtmlSanitizer.new(kramdown_doc.to_html).sanitize(allowed_elements: @allowed_elements)
-                            else
-                              kramdown_doc.to_html
-                            end
+        html = if @options[:sanitize]
+                 HtmlSanitizer.new(kramdown_doc.to_html).sanitize(allowed_elements: @allowed_elements)
+               else
+                 kramdown_doc.to_html
+               end
 
-                     Govspeak::PostProcessor.process(html, self)
-                   end
+        unless @footnote_definition_html.nil?
+          regex = /<div class="footnotes".*[<\/div>]/m
+
+          if html.scan(regex).empty?
+            html << @footnote_definition_html
+          else
+            html.gsub!(regex, @footnote_definition_html)
+          end
+        end
+
+        Govspeak::PostProcessor.process(html, self)
+      end
     end
 
     def to_liquid
       to_html
     end
@@ -108,18 +120,55 @@
     end
 
     def preprocess(source)
       source = Govspeak::BlockquoteExtraQuoteRemover.remove(source)
       source = remove_forbidden_characters(source)
+
+      legislative_list_footnote_definitions(source)
+
       self.class.extensions.each do |_, regexp, block|
         source.gsub!(regexp) do
           instance_exec(*Regexp.last_match.captures, &block)
         end
       end
       source
     end
 
+    def legislative_list_footnote_definitions(source)
+      is_legislative_list = source.scan(/\$LegislativeList.*?\[\^\d\]*.*?\$EndLegislativeList/m).size.positive?
+      footnotes = source.scan(/\[\^(\d)\]:(.*)/)
+      @acronyms = source.scan(/(?<=\*)\[(.*)\]:(.*)/)
+
+      if is_legislative_list && footnotes.size.positive?
+        list_items = footnotes.map do |footnote|
+          number = footnote[0]
+          text = footnote[1].strip
+          footnote_definition = Govspeak::Document.new(text).to_html[/(?<=<p>).*(?=<\/p>)/]
+
+          <<~HTML_SNIPPET
+            <li id="fn:#{number}" role="doc-endnote">
+              <p>
+                #{footnote_definition}<a href="#fnref:#{number}" class="reversefootnote" role="doc-backlink" aria-label="go to where this is referenced">↩</a>
+              </p>
+            </li>
+          HTML_SNIPPET
+        end
+
+        @footnote_definition_html = <<~HTML_CONTAINER
+          <div class="footnotes" role="doc-endnotes">
+            <ol>
+              #{list_items.join.strip}
+            </ol>
+          </div>
+        HTML_CONTAINER
+      end
+
+      unless @footnote_definition_html.nil? && @acronyms.size.positive?
+        add_acronym_alt_text(@footnote_definition_html)
+      end
+    end
+
     def remove_forbidden_characters(source)
       # These are characters that are not deemed not suitable for
       # markup: https://www.w3.org/TR/unicode-xml/#Charlist
       source.gsub(Sanitize::REGEX_UNSUITABLE_CHARS, "")
     end
@@ -159,11 +208,11 @@
         \] # match end of link markdown
         \( # match start of link text markdown
           ([^)]+)  # capture inside of link text markdown
         \) # match end of link text markdown
         \s*  # any whitespace between opening bracket and link
-      {\/button} # match ending bracket
+      {/button} # match ending bracket
       (?:\r|\n|$) # non-capturing match to make sure end of line and linebreak
     }x) do |attributes, text, href|
       button_classes = "govuk-button"
       /cross-domain-tracking:(?<cross_domain_tracking>.[^\s*]+)/ =~ attributes
       data_attribute = ""
@@ -293,10 +342,22 @@
       Govspeak::KramdownOverrides.with_kramdown_ordered_lists_disabled do
         Kramdown::Document.new(body.strip).to_html.tap do |doc|
           doc.gsub!("<ul>", "<ol>")
           doc.gsub!("</ul>", "</ol>")
           doc.sub!("<ol>", '<ol class="legislative-list">')
+
+          footnotes = body.scan(/\[\^(\d+)\]/).flatten
+
+          footnotes.each do |footnote|
+            html = "<sup id=\"fnref:#{footnote}\" role=\"doc-noteref\">" \
+            "<a href=\"#fn:#{footnote}\" class=\"footnote\" rel=\"footnote\">" \
+            "[footnote #{footnote}]</a></sup>"
+
+            doc.sub!(/(\[\^#{footnote}\])/, html)
+          end
+
+          add_acronym_alt_text(doc) if @acronyms.size.positive?
         end
       end
     end
 
     extension("numbered list", /^[ \t]*((s\d+\.\s.*(?:\n|$))+)/) do |body|
@@ -383,9 +444,15 @@
       @kramdown_doc ||= Kramdown::Document.new(preprocess(@source), @options)
     end
 
     def encode(text)
       HTMLEntities.new.encode(text)
+    end
+
+    def add_acronym_alt_text(html)
+      @acronyms.each do |acronym|
+        html.gsub!(acronym[0], "<abbr title=\"#{acronym[1].strip}\">#{acronym[0]}</abbr>")
+      end
     end
   end
 end
 
 I18n.load_path.unshift(