knownConditionTrueFalse not triggered when global variable is involved

[Hubert Melchert]

I recently discovered a bug in my code that cppcheck hasn't warned me about. In my opinion cppcheck should be fully aware of this error . In simpified provided example below cppcheck is silent when it comes to if condition being always false:

typedef enum
{
    testEnum_a = 1,
    testEnum_b = 2
}testEnum_t;

testEnum_t e = 3;

int main(void)
{
    if(e & testEnum_a & testEnum_b)
    {
        printf("True\n");
    }
    else
    {
        printf("False\n");
    }

    return 0;
}

If i remove variable "e" from that condition cppcheck warns me about always false condition. When I prepend definition of "e" with static cppcheck works as expected.

static testEnum_t e = 3;

int main(void)
{
    if(e & testEnum_a & testEnum_b)
    {
        printf("True\n");
    }
    else
    {
        printf("False\n");
    }

    return 0;
}

cppcheck --enable=all test.c
Checking test.c ...
test.c:15:23: style: Condition 'e&testEnum_a&testEnum_b' is always false [knownConditionTrueFalse]
    if(e & testEnum_a & testEnum_b)
                      ^

When I move defition of "e" to main cppcheck detects always false condition.
So why cppcheck can't warn about knowConditionTrueFalse when there is a global variable with external linkage involved? Regardless of actual value of "e" condition will always be false.

[versat]

You are right, testEnum_a & testEnum_b will always be false, so Cppcheck could warn even if the value of e is not known. Maybe some check to avoid false positives avoids this warning to be generated, but I do not know. I think this false negative is worth a ticket.

[versat]

While trying to reproduce the issue and finding a minimal example I found out something.
When the if condition is changed to testEnum_a & testEnum_b & e the issue is reported:

$ ./cppcheck --enable=style forum.c
Checking forum.c ...
forum.c:11:33: style: Condition 'testEnum_a&testEnum_b&e' is always false [knownConditionTrueFalse]
    if( testEnum_a & testEnum_b & e)
                                ^

So it matters where exactly the unknown/unsure variable value is in the condition. But in this case that does not make a difference and the issue should be detected regardless of the order of the variables in the condition.

Edit: I have created a ticket: https://trac.cppcheck.net/ticket/9527
Thanks for reporting that issue.

[Hubert Melchert]

Can you create a ticket based on this post? If not then I will create a Trac account and report this.

[versat]

I have created a ticket: https://trac.cppcheck.net/ticket/9527