Typically with
```
int F(int *f);
void F2(int *a, int *b) {
int c = *a;
F(a); // modifies *a
if (b && c != *a) {}
}
```
we would get the following FP:
```
[test.cpp:3] -> [test.cpp:5]: (style) The comparison 'c != *a' is always false because 'c' and '*a' represent the same value.\n
```
I guess it boils down to isSameExpression only checking that the
expression is the same (in the above case, "*a" and "*a" are indeed the
same), but there's not real check on the values.
So the patch here is a bit hackish, and we still have false negatives in
cases with dereferenced pointers.
* Refactor isNullOperand out of FwdAnalysis
* Improve isNullOperand
* Fix redundantAssignment FP with unsigned zero
* isNullValue check number
* Enhance isNullOperand to handle c++ casts
Also handle cast of NULL.
* Add cases for 9356
* 9356: Prevent false positive when passing non-const reference to member constructor
This workarounds false positives 'Parameter can be declared with const [constParameter]'
when said parameter is used in constructor call. It assume the
constructor call might change the parameter (without any checks.
The drawback is that we have false negative, in cases where we could
check the constructor actually takes a const reference, or a copied by
value parameter.
* Add todo comment in isVariableMutableInInitializer
Previously, cppcheck discarded the `extern "C"` specifier. This patch modifies cppcheck to parse each as a Scope in the symbol database, then uses that scope to avoid false positives when making recommendations about changing a function argument to be a reference (since variable references is a C++ feature, unavailable in C, and thus unavailable in `extern "C"`).
* use range loops
* removed redundant string initializations
* use nullptr
* use proper boolean false
* removed unnecessary continue from end of loop
* removed unnecessary c_str() usage
* use emplace_back()
* removed redundant void arguments
* Add impossible category
* Replace values
* Try to adjust known values
* Add ! for impossible values
* Add impossible with possible values
* Remove contradictions
* Add values when the branch is not dead
* Only copy possible values
* Dont bail on while loops
* Load std lib in valueflow
* Check for function calls
* Fix stl errors
* Fix incorrect impossible check
* Fix heap-after-use error
* Remove impossible values when they are lowered
* Show the bound and remove overlaps
* Infer conditions
* Dont push pointer values through dynamic_cast
* Add test for dynamic_cast issue
* Add shifttoomanybits test
* Add test for div by zero
* Add a test for issue 9315
* Dont make impossible value inconclusive
* Fix FP with shift operator
* Improve handleKnownValuesInLoop for impossible values
* Fix cppcheck warning
* Fix impossible values for ctu
* Bailout for streams
* Check equality conditions
* Fix overflows
* Add regression test for 9332
* Remove duplicate conditions
* Skip impossible values for invalid value
* Check for null
* Rename bound to range
* Formatting
Building with enhanced clang warnings indicated a large number of
instances with the warning:
`warning: zero as null pointer constant`
Recommended practice in C++11 is to use `nullptr` as value for
a NULL or empty pointer value. All instances where this warning
was encountered were corrected in this commit.
Where warning was encountered in dependency code (i.e. external library)
no chnages were made. Patching will be offered upstream.
This uses the lifetime analysis to check when comparing pointer that point to different objects:
```cpp
int main(void)
{
int foo[10];
int bar[10];
int diff;
if(foo > bar) // Undefined Behavior
{
diff = 1;
}
return 0;
}
```
This reworks constStatement to find more issues. It catches issue [8827](https://trac.cppcheck.net/ticket/8827):
```cpp
extern void foo(int,const char*,int);
void f(int value)
{
foo(42,"test",42),(value&42);
}
```
It also catches from issue [8451](https://trac.cppcheck.net/ticket/8451):
```cpp
void f1(int x) {
1;
(1);
(char)1;
((char)1);
!x;
(!x);
~x;
}
```
And also:
```cpp
void f(int x) {
x;
}
```
The other examples are not caught due to incomplete AST.
Add a call to simplifyPlatformTypes() in
SymbolDatabase::setValueTypeInTokenList() to simplify return types of
library configured functions. This fixes the FN in #8141. Regression
tests are added, both for the original issue and another FN in the comments.
In order to do that, move simplifyPlatformTypes() to TokenList from Tokenizer.
This is a pure refactoring and does not change any behaviour. The code was
literally copy-pasted from one file to another and in two places
'list.front()' was changed to 'front()'.
When adding the call to simplifyPlatformTypes(), the original type of
v.size() where v is a container is changed from 'size_t' to 'std::size_t'.
Tests are updated accordingly. It can be noted that if v is declared as
'class fred : public std::vector<int> {} v', the original type of 'v.size()'
is still 'size_t' and not 'std::size_t'.
* Remove newlines after check(
* Remove unneeded statements after if-statements
As an example, the previous test case
check(
"bool foo(int x) {\n"
" if (x < 0)"
" return true;\n"
" return false;\n"
"}");
is changed to
check("void foo(int x) {\n"
" if (x < 0) {}\n"
"}");
The unsigned less than zero checker looked for patterns like "<= 0".
Switching to use valueflow improves the checker in a few aspects.
First, it removes false positives where instead of 0, the code is using
0L, 0U, etc. Instead of having to hard code the different variants of 0,
valueflow handles this automatically. This fixes FPs on the form
uint32_t value = 0xFUL;
void f() {
if (value < 0u)
{
value = 0u;
}
}
where 0u was previously not recognized by the checker. This fixes#8836.
Morover, it makes it possible to handle templates properly. In commit
fa076598ad, all warnings inside templates
were made inconclusive, since the checker had no idea if "0" came from
a template parameter or not.
This makes it possible to not warn for the following case which was
reported as a FP in #3233
template<int n> void foo(unsigned int x) {
if (x <= n);
}
foo<0>();
but give a warning for the following case
template<int n> void foo(unsigned int x) {
if (x <= 0);
}
Previously, both these cases gave inconclusive warnings.
Finally, it makes it possible to give warnings for the following code:
void f(unsigned x) {
int y = 0;
if (x <= y) {}
}
Also, previously, the checker for unsigned variables larger than 0, the
checker used the string of the astoperand. This meant that for code like
the following:
void f(unsigned x, unsigned y) {
if (x -y >= 0) {}
}
cppcheck would output
[unsigned-expression-positive.c] (style) Unsigned variable '-' can't be negative so it is unnecessary to test it.
using expressionString() instead gives a better error message
[unsigned-expression-positive.c] (style) Unsigned expression 'x-z' can't be negative so it is unnecessary to test it.
* Add regression test for #6906
Ticket #6906 was fixed in f65cf220ba.
Add a test to make sure there are no regressions.
* Add regression test for #7284
Ticket #7284 was fixed in 5d1fdf7958.
Add tests to avoid regressions.
Add a check for function arguments that can be constant:
```cpp
extern void bar(int);
void f(int x) {
bar((x & 0x01) >> 7); // function 'bar' is always called with a '0'-argument
}
```
The while part of a do-while loop looks almost like a function call, so
extend the check for function calls to ignore while-statements.
Note that there was only an FP when checking c-code, since the check is
disabled for c++-code. Therefore, make sure the test cases are run on a
c-file.
* Follow variables when comparing same expression
* Remove assert include
* Dont follow function arguments
* Improve the checking to check more cases
* Add more tests
* Check if the variable is used inside a loop
* Follow both variables
* Only skip loops when variable is modified in scope
* Fix FP when followed variable is modified
* Dont follow arrays
* Skip pointer indirection
* Make recursive
* Improve checking more variables
* Fix test with sizeof
* Skip following operators
* Fix test when using sizeof
* Dont check every step
* Use early returns
* Update test to use a loop instead of conditional
* Add static
* Check variables are global
* Check local variables in another scope
* Fix issue with const pointers
* Distinguish between pointer indirection and multiply
* Use simple match
* Prevent crash with uniform initialization
* Use unary op and ast to detect pointer indirection
* Expand error message when expression do not match exactly
* Add errorpath to issameexpression
* Revert "Clarify warning message for 'Same expression on both sides of operator'"
This reverts commit 0e491b41a8.
* Check if the tokens are the same
* Report the operator and not the expressions
* Check for duplicate assignments
* Improve checking of expression
* Add more tests
* Use simple match
* Improve robustness of check
* check for null
* Reduce side effects by checking for side effects
* Improve verbose message
* Reword the error message
A unique pointer should only be passed by value or by const reference.
Passing by value means transferring ownership it makes no sense to have a pass by value Warning for unique_ptr
Signed-off-by: Andreas Pokorny <andreas.pokorny@siemens.com>
Struct arguments are either pushed to the stack or passed in the
registers. Since both methods operate on machine words it is logical to
tie the maximum size of a struct argument that doesn't trigger
passedByValue diagnostic to the size of the machine word.
Also guessed size of STL classes is set to 3*sizeof_pointer, this better
represents reality and ensures that structs containing them will still
trigger passedByValue.
Daniel Marjamäki
Paul Fultz II
shaneasd
Ken-Patrick Lehrmann
miltolstoy
PKEuS
amai2012
Rikard Falkeborn
Dmitry-Me
Steve Mokris
Oliver Stöneberg
Scott Furry
IOBYTE
orbitcowboy
lordylike
amai
Matthias Krüger
Andreas Pokorny
Stas Cymbalov
Ayaz Salikhov
uburuntu
PKEuS