static inline bool ParseTimestampISO8601()

in cpp/src/arrow/util/value_parsing.h [661:789]
91 lines of code
43 McCabe index (conditional complexity)

static inline bool ParseTimestampISO8601(const char* s, size_t length,
                                         TimeUnit::type unit, TimestampType::c_type* out,
                                         bool* out_zone_offset_present = NULLPTR) {
  using seconds_type = std::chrono::duration<TimestampType::c_type>;

  // We allow the following zone offset formats:
  // - (none)
  // - Z
  // - [+-]HH(:?MM)?
  //
  // We allow the following formats for all units:
  // - "YYYY-MM-DD"
  // - "YYYY-MM-DD[ T]hhZ?"
  // - "YYYY-MM-DD[ T]hh:mmZ?"
  // - "YYYY-MM-DD[ T]hh:mm:ssZ?"
  //
  // We allow the following formats for unit == MILLI, MICRO, or NANO:
  // - "YYYY-MM-DD[ T]hh:mm:ss.s{1,3}Z?"
  //
  // We allow the following formats for unit == MICRO, or NANO:
  // - "YYYY-MM-DD[ T]hh:mm:ss.s{4,6}Z?"
  //
  // We allow the following formats for unit == NANO:
  // - "YYYY-MM-DD[ T]hh:mm:ss.s{7,9}Z?"
  //
  // UTC is always assumed, and the DataType's timezone is ignored.
  //

  if (ARROW_PREDICT_FALSE(length < 10)) return false;

  seconds_type seconds_since_epoch;
  if (ARROW_PREDICT_FALSE(!ParseYYYY_MM_DD(s, &seconds_since_epoch))) {
    return false;
  }

  if (length == 10) {
    *out = util::CastSecondsToUnit(unit, seconds_since_epoch.count());
    return true;
  }

  if (ARROW_PREDICT_FALSE(s[10] != ' ') && ARROW_PREDICT_FALSE(s[10] != 'T')) {
    return false;
  }

  if (out_zone_offset_present) {
    *out_zone_offset_present = false;
  }

  seconds_type zone_offset(0);
  if (s[length - 1] == 'Z') {
    --length;
    if (out_zone_offset_present) *out_zone_offset_present = true;
  } else if (s[length - 3] == '+' || s[length - 3] == '-') {
    // [+-]HH
    length -= 3;
    if (ARROW_PREDICT_FALSE(!detail::ParseHH(s + length + 1, &zone_offset))) {
      return false;
    }
    if (s[length] == '+') zone_offset *= -1;
    if (out_zone_offset_present) *out_zone_offset_present = true;
  } else if (s[length - 5] == '+' || s[length - 5] == '-') {
    // [+-]HHMM
    length -= 5;
    if (ARROW_PREDICT_FALSE(!detail::ParseHHMM(s + length + 1, &zone_offset))) {
      return false;
    }
    if (s[length] == '+') zone_offset *= -1;
    if (out_zone_offset_present) *out_zone_offset_present = true;
  } else if ((s[length - 6] == '+' || s[length - 6] == '-') && (s[length - 3] == ':')) {
    // [+-]HH:MM
    length -= 6;
    if (ARROW_PREDICT_FALSE(!detail::ParseHH_MM(s + length + 1, &zone_offset))) {
      return false;
    }
    if (s[length] == '+') zone_offset *= -1;
    if (out_zone_offset_present) *out_zone_offset_present = true;
  }

  seconds_type seconds_since_midnight;
  switch (length) {
    case 13:  // YYYY-MM-DD[ T]hh
      if (ARROW_PREDICT_FALSE(!detail::ParseHH(s + 11, &seconds_since_midnight))) {
        return false;
      }
      break;
    case 16:  // YYYY-MM-DD[ T]hh:mm
      if (ARROW_PREDICT_FALSE(!detail::ParseHH_MM(s + 11, &seconds_since_midnight))) {
        return false;
      }
      break;
    case 19:  // YYYY-MM-DD[ T]hh:mm:ss
    case 21:  // YYYY-MM-DD[ T]hh:mm:ss.s
    case 22:  // YYYY-MM-DD[ T]hh:mm:ss.ss
    case 23:  // YYYY-MM-DD[ T]hh:mm:ss.sss
    case 24:  // YYYY-MM-DD[ T]hh:mm:ss.ssss
    case 25:  // YYYY-MM-DD[ T]hh:mm:ss.sssss
    case 26:  // YYYY-MM-DD[ T]hh:mm:ss.ssssss
    case 27:  // YYYY-MM-DD[ T]hh:mm:ss.sssssss
    case 28:  // YYYY-MM-DD[ T]hh:mm:ss.ssssssss
    case 29:  // YYYY-MM-DD[ T]hh:mm:ss.sssssssss
      if (ARROW_PREDICT_FALSE(!detail::ParseHH_MM_SS(s + 11, &seconds_since_midnight))) {
        return false;
      }
      break;
    default:
      return false;
  }

  seconds_since_epoch += seconds_since_midnight;
  seconds_since_epoch += zone_offset;

  if (length <= 19) {
    *out = util::CastSecondsToUnit(unit, seconds_since_epoch.count());
    return true;
  }

  if (ARROW_PREDICT_FALSE(s[19] != '.')) {
    return false;
  }

  uint32_t subseconds = 0;
  if (ARROW_PREDICT_FALSE(
          !detail::ParseSubSeconds(s + 20, length - 20, unit, &subseconds))) {
    return false;
  }

  *out = util::CastSecondsToUnit(unit, seconds_since_epoch.count()) + subseconds;
  return true;
}