CGI::Escape#unescape_html

# File tmp/rubies/ruby-4.1.0-preview1/lib/cgi/escape.rb, line 113
def unescapeHTML(string)
  enc = string.encoding
  unless enc.ascii_compatible?
    if enc.dummy?
      origenc = enc
      enc = Encoding::Converter.asciicompat_encoding(enc)
      string = enc ? string.encode(enc) : string.b
    end
    string = string.gsub(Regexp.new('&(apos|amp|quot|gt|lt|#[0-9]+|#x[0-9A-Fa-f]+);'.encode(enc))) do
      case $1.encode(Encoding::US_ASCII)
      when 'apos'                then "'".encode(enc)
      when 'amp'                 then '&'.encode(enc)
      when 'quot'                then '"'.encode(enc)
      when 'gt'                  then '>'.encode(enc)
      when 'lt'                  then '<'.encode(enc)
      when /\A#0*(\d+)\z/        then $1.to_i.chr(enc)
      when /\A#x([0-9a-f]+)\z/i  then $1.hex.chr(enc)
      end
    end
    string.encode!(origenc) if origenc
    return string
  end
  return string unless string.include? '&'
  charlimit = case enc
              when Encoding::UTF_8; 0x10ffff
              when Encoding::ISO_8859_1; 256
              else 128
              end
  string = string.b
  string.gsub!(/&(apos|amp|quot|gt|lt|\#[0-9]+|\#[xX][0-9A-Fa-f]+);/) do
    match = $1
    case match
    when 'apos'                then "'"
    when 'amp'                 then '&'
    when 'quot'                then '"'
    when 'gt'                  then '>'
    when 'lt'                  then '<'
    else
      # Numeric character reference. Decode into the binary buffer so that a
      # non-ASCII byte already present in it never triggers an encoding
      # compatibility error on concatenation; the trailing force_encoding
      # re-tags the whole buffer. This mirrors the C extension's
      # optimized_unescape_html.
      n = match.start_with?('#x', '#X') ? match[2..-1].hex : match[1..-1].to_i
      if n >= charlimit
        "&#{match};"            # out of range: keep the reference verbatim
      elsif charlimit > 256
        [n].pack('U').b         # UTF-8: code point bytes, surrogates included (like rb_enc_mbcput)
      else
        n.chr.b                 # ISO-8859-1 / ASCII: single byte
      end
    end
  end
  string.force_encoding enc
end