Fix expected output in TestCondition::alwaysTrue and TestCondition::alwaysTrueContainer

[andreas-schwab]

No description provided.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[chrchr-github]

Can this be tested using e.g. -funsigned-char?

[andreas-schwab]

-funsigned-char may change the ABI and thus invokes undefined behaviour.

[chrchr-github]

I was actually thinking of cppcheck's own -funsigned-char switch.

But I see now that we only run a single test (TestSymbolDatabase) in build_uchar (CI-unixish) with unsigned char, so the bad output was not detected. I wonder if we should run all tests?

[danmar]

I wonder if we should run all tests?

it sounds reasonable to me.

however is that related to this PR? the std::numeric_limits<char>::is_signed would still be true no matter which cppcheck options is used?

[chrchr-github]

I wonder if we should run all tests?

it sounds reasonable to me.

however is that related to this PR? the std::numeric_limits<char>::is_signed would still be true no matter which cppcheck options is used?

Yes, -funsigned-charprobably doesn't help here. I hadn't realized that we have build_uchar in the CI, so we should run all tests there.

[firewave]

Different output depending on build flags feels like a bug to me. It should be platform-agnostic and this should rather be handled in the check than the test.