Method access question

[engaarea]

Hi all,
How does this work?

class a{
public:
int geta(){
return 2;
}
};
int main(){
a *ptr = 0;
cout<<"Value :"<<ptr->geta();
}

this prints
Value: 100

I expected a runtime error. but this worked just fine. How?

Regards
area

[Ivan Vecerina]

<en******@gmail.comwrote in message
news:11*********************@h48g2000cwc.googlegro ups.com...
: Hi all,
: How does this work?
:
: class a{
: public:
: int geta(){
: return 2;
: }
: };
:
:
: int main(){
: a *ptr = 0;
: cout<<"Value :"<<ptr->geta();
: }
:
: this prints
: Value: 100
:
: I expected a runtime error. but this worked just fine. How?
"worked just fine" ?? I assume there is a typo in your
post then (2 vs 100).

Anyway:
- The C++ standard does not require any checks
for NULL or invalid pointers during dereferencing.
What is specified in that case is "undefined behavior"
(i.e. anything could happen).
However, some C++ platfomrs may offer the possibility
to have automated runtime checks for such errors.
- In the above example, the executed code does not
actually need to access the *ptr object to execute
function geta() -- because no data member is accessed,
and geta() is *not* declared as virtual. So although
the behavior of the code is formally undefined, it
will just happen to run without error on must C++ platforms.
hth --Ivan
--
http://ivan.vecerina.com/contact/?subject=NG_POST <- email contact form
Brainbench MVP for C++ <http://www.brainbench.com

[Ron Natalie]

Ivan Vecerina wrote:
So although

the behavior of the code is formally undefined, it
will just happen to run without error on must C++ platforms.

Yep, that's one of the insidious possibilities of undefined
behavior. It may work apparently without error... until some
day when something else changes (like you update to the next
version of the compiler) and BOOM! code suddenly breaks.