Issue with Timestamp Parsing

[171930433]

$GPGGA,092750.000,5321.6802,N,00630.3372,W,1,8,1.03,61.7,M,55.2,M,,76
$GPRMC,092750.000,A,5321.6802,N,00630.3372,W,0.02,31.66,280511,,,A43

Here is a summary of the issues I've encountered and a suggestion:

1. When I configure the library to parse only GPRMC for timestamping (e.g., for PPS synchronization), the time-of-day fields are not parsed and are therefore inaccessible.
2. Even with GPGGA enabled, which does provide access to time, the library discards the fractional part of the seconds, losing all sub-second precision.
3. I am not using the LWGPS_CFG_STATEMENT_PUBX_TIME option as it is for a non-standard, vendor-specific sentence.

Suggestion: For improved clarity, would it be better if the data structure for parsed statements more directly mapped the fields from each sentence type? For example, using a single unified struct for all common data (time, date, position, etc.) that each parser populates, rather than separate structs in a union.

Best regards,

[MaJerle]

Can you do a pull request with your proposed changes? The original objective was to make it small and lightweight, therefore modular.

We can conditionally enable the time fields when RMC or GGA are enabled, these lines:

lwgps/lwgps/src/include/lwgps/lwgps.h

Lines 122 to 132 in 5e79239

	#if LWGPS_CFG_STATEMENT_GPRMC || __DOXYGEN__
	/* Information related to GPRMC statement */
	uint8_t is_valid; /*!< GPS valid status */
	lwgps_float_t speed; /*!< Ground speed in knots */
	lwgps_float_t course; /*!< Ground coarse */
	lwgps_float_t variation; /*!< Magnetic variation */
	uint8_t date; /*!< Fix date */
	uint8_t month; /*!< Fix month */
	uint8_t year; /*!< Fix year */
	uint8_t date_valid; /*!< Flag whether date data has been provided by GPS modem in a valid `6` digit format */
	#endif /* LWGPS_CFG_STATEMENT_GPRMC || __DOXYGEN__ */

And parse time either on:

• RMC, if enabled
• GGA, if enabled, and RMC not enabled

We then guarantee that the time is always parsed if one of them is available and we avoid parsing it on both.