void encode_double()

in src/codec/number.cpp [82:102]

• 12 lines of code
• 1 McCabe index (conditional complexity)

void encode_double(encode_context &context, double value) {
  // The maximum buffer size required to emit a double in base 10, for decimal
  // and exponential representations, is 25 bytes; based on the settings used
  // below for the DoubleToStringConverter. We add another byte for the null
  // terminator, but it is not actually needed because we don't finalize the
  // builder.
  const auto max_required_size = 26;
  const auto p = reinterpret_cast<char *>(context.reserve(max_required_size));

  // The converter is based on the ECMAScript converter, but will not convert
  // special values, like Infinity and NaN, since JSON does not support those.
  using dtoa_converter = double_conversion::DoubleToStringConverter;
  const dtoa_converter converter(
      dtoa_converter::UNIQUE_ZERO | dtoa_converter::EMIT_POSITIVE_EXPONENT_SIGN,
      nullptr, nullptr, 'e', -6, 21, 6, 0);

  using dtoa_builder = double_conversion::StringBuilder;
  dtoa_builder builder(p, max_required_size);
  detail::fail_if(context, !converter.ToShortest(value, &builder), "Special values like 'Infinity' or 'NaN' are supported in JSON.");
  context.advance(builder.position());
}