invalid use of Trice0 will misbehave trice insert

[KammutierSpule]

Trice0( variable ); // invalid use
Trice0( "text %u", variable ); // invalid use

Will cause that the next Trices will be ignored or not updated.
(also if Trice0 are // commented)

See also: #464

[rokath]

Am I understand you right, that the first line causes the trice tool not to insert further IDs in that file? That's indeed an issue. Unfortunately I am not a parser expert and the trice tool is somehow limited.

There is probably no quick fix right now. See also comment in https://github.com/rokath/trice/blob/master/internal/id/insertIDs.go near line 280.

[KammutierSpule]

also, it happens with clean too.
If there is an invalid use of Trice0, trice clean will not be able to clean next ids on the file

[rokath]

Yes, it is the same parser. I am going to add a chapter about that to https://github.com/rokath/trice/blob/master/docs/TriceUserGuide.md#9-additional-hints.

[rokath]

It is added now into https://github.com/rokath/trice/blob/master/docs/TriceUserGuide.md#922-limited-trice-parser-capabilities.

[rokath]

The following sequence inside trice/examples/OpenCM3_STM32F411_Nucleo/main.c

                // Depending on mode, either print this string to
		// UART (mode 0), or the Trice write buffer (mode 200).
		TRICE( Id( 4030),"Hello, TRICE, %d\n", 42);

		// TRICE() with a string parameter only is problematic.
		// See discussion on https://github.com/rokath/trice/issues/279
		// TRICE0() works in either case
		#ifdef __STRICT_ANSI__
		// if compiled with e.g. --std=c99
		TRICE0( Id( 4683),"Hello, TRICE\n");
		#else
		TRICE( Id( 2926),"Hello, TRICE\n");
		TRICE0( Id( 7263),"Hello, TRICE0()\n");
		#endif

gets changed into

		// Depending on mode, either print this string to
		// UART (mode 0), or the Trice write buffer (mode 200).
		TRICE(Id(0), "Hello, TRICE, %d\n", 42);

		// TRICE(Id(0), "Hello, TRICE\n");
		#else
		TRICE(Id(0), "Hello, TRICE\n");
		TRICE0(Id(0), "Hello, TRICE0()\n");
		#endif

so these lines are disappearing:

		// TRICE() with a string parameter only is problematic.
		// See discussion on https://github.com/rokath/trice/issues/279
		// TRICE0() works in either case
		#ifdef __STRICT_ANSI__
		// if compiled with e.g. --std=c99

when running renewIDs_in_examples_and_test_folder.sh in branch devel. This needs a special test first.

[rokath]

I propose to close this issue for less issue noise. Normal usage is trice(... and all these cases are evaluated for matching value counts. If a user writes TRICE0(... or TRICE_3(... incorrectly, at least the compiler will complain. Also this limitation is documented now.