The analyzer has detected a situation where a variable is initialized or assigned a new value and is expected to be checked in the condition of a subsequent 'if' statement but another variable is checked instead.
This error is demonstrated by the following example:
int ret = syscall(....);
if (ret != -1) { .... }
....
int ret2 = syscall(....);
if (ret != -1) { .... } // <=
Programmers often need to check the value returned by a function but use a wrong variable name in the condition of the 'if' statement. This mistake is typically made when you clone a code fragment but forget to modify the name of the variable in the condition. In the example above, the programmer forgot to change the name 'ret' to 'ret2'.
Fixed version:
int ret2 = syscall(....);
if (ret2 != -1) { .... }
The following example also demonstrates this mistake:
obj->field = ....;
if (field) ....;
Both the variable and the class member have the same name, which makes it easy to confuse one with the other.
This diagnostic is heuristic; it splits the names of variables into individual strings and compares the respective parts to conclude if there is a typo. It also performs a basic type check, aiming at reducing the number of false positives.
The diagnostic may often be triggered by code like this:
var->m_abc = ....;
var->m_cba = ....;
if (var->m_abc) // <=
{
....
}
Fragments like this are usually correct. You can either suppress such warnings or swap the assignments so that the variable to be checked is assigned a value immediately before the 'if' statement:
var->m_cba = ....;
var->m_abc = ....;
if (var->m_abc)
{
....
}
Keeping the assignment and the check close to each other also makes the code more readable.
This diagnostic is classified as:
You can look at examples of errors detected by the V1051 diagnostic. |