Unit test fails when VER_2 headers are used

[skliper]

Describe the bug
Event test strings are different when the MsgId size changes, but test looks for explict message:

sample_app/fsw/src/sample_app.c

Lines 274 to 277 in c1348d6

	CFE_EVS_SendEvent(SAMPLE_INVALID_MSGID_ERR_EID,
	CFE_EVS_EventType_ERROR,
	"SAMPLE: invalid command packet,MID = 0x%x",
	(unsigned int)CFE_SB_MsgIdToValue(MsgId));

sample_app/unit-test/coveragetest/coveragetest_sample_app.c

Line 290 in c1348d6

UT_CheckEvent_Setup(&EventTest, SAMPLE_INVALID_MSGID_ERR_EID, "SAMPLE: invalid command packet,MID = 0xffff");

sample_app/unit-test/coveragetest/coveragetest_sample_app.c

Line 549 in c1348d6

UT_CheckEvent_Setup(&EventTest, SAMPLE_LEN_ERR_EID, "Invalid Msg length: ID = 0xFFFF, CC = 0, Len = 18, Expected = 8");

!VER_2 = 0xffff, VER_2 =0xffffffff (or capitalized based on the format string)

To Reproduce
Set MESSAGE_FORMAT_IS_CCSDS_VER_2, build and run unit tests.

Expected behavior
Unit tests should pass with either configuration

Code snips
See above.

System observed on:

• Hardware: cFS Dev VM
• OS: Ubuntu 18.04
• Versions: main bundle

Additional context
None

Reporter Info
Jacob Hageman - NASA/GSFC

[skliper]

Probably easiest to just remove the explicit string check (pass in blank).

[jphickey]

Yeah, this is why I don't like to check derived output in UT. My recommendation is (and has always been) to only check the direct outputs of the unit under test, which in this case is the format string (i.e. "SAMPLE: invalid command packet,MID = 0x%x") not the fully formed output string, which is dependent on a number of factors outside the control of the unit under test.

IIRC there were some folks specifically wanting to do this type of test and hence why it was included in sample_app example.

A possible compromise would be to only check the format string.

[skliper]

Note with nasa/cFE#880 the reported error is now consistent since message ID base type is uint32.