A Simple C++ Question.

Sivaraman Dhamodharan

Have a look at the below code:

#include "stdafx.h"
#include "conio.h"

int _tmain(int argc, _TCHAR* argv[])
{
//Snippet 1
char name[20] = "Some Name";
printf("%s\n", name);
name[1] = 'a';
printf("%s\n", name);

//Snippet 2
char \* name2 = "Some Name";
printf("%s\\n", name2);
name2\[1\] = 'a';
printf("%s\\n", name2);

getch();
return 0;

}

Why we get Runtime error when we modify name2[1] = 'a';?

Programming Article

David Crow

It all boils down to the difference between an array and a pointer. For name, you have set aside room for 20 characters (that are initialized to "Some Name"). For name2, you are pointing to a static piece of memory. That memory cannot be changed.

"One man's wage rise is another man's price increase." - Harold Wilson

"Fireproof doesn't mean the fire will never come. It means when the fire comes that you will be able to withstand it." - Michael Simmons

"Show me a community that obeys the Ten Commandments and I'll show you a less crowded prison system." - Anonymous

Albert Holguin

In this case... your definition of name2 is what is referred to as a string literal and points to data that is considered constant. See here[^].

Maximilien

You are trying to modify a const char array (pointer). This is not C++. if you were writing proper C++, you would not get those problems (or a lot less likely).

std::string name = "some name";
std::cout << name << std::endl;

std::string name2 = "some name";
std::cout << name2 << std::endl;

name2[1] = 'a'; // need to check that the index is within the string size.
std::cout << name2 << std::endl;

Watched code never compiles.

Chuck OToole

Technically, to pick nits, it *is* C++ as C++ is a superset of C. What you are really saying is that it isn't "Object Oriented Programming" which is true, but it is still within the definition of C++.

Albert Holguin

Agree with Chuck... technically it is C++.

jschell

Chuck O'Toole wrote:

Technically, to pick nits, it *is* C++ as C++ is a superset of C

Does the newest ANSI C++ standard incorporate the latest ANSI C standard as a subset? Last I heard that was not the case

Aescleal

C++98 is not a superset of C90 - especially when you consider const and what happens when you cast away const. C allows modification of literals (string and constants) through pointers while C++ doesn't or rather considers it undefined behaviour.

Chuck OToole

The point is, C++ did *not* decommit the "char" data type nor pointers. std::string is *not* the only way to do string manipulation in C++. It may be the perferred way for some people but it's not the only way.

Chuck OToole

The point is, C++ did *not* decommit the "char" data type nor pointers. std::string is *not* the only way to do string manipulation in C++. It may be the perferred way for some people but it's not the only way.

jschell

Chuck O'Toole wrote:

The point is, C++ did *not* decommit the "char" data type nor pointer

I was responding to what appeared to be a general comment about the language in general and in its entirety and not just one small part. And that general comment, at this time (new standards), is wrong.

Chuck OToole

Context dude, gotta read comments in the context of the thread

jschell

Chuck O'Toole wrote:

Context dude, gotta read comments in the context of the thread

And respond in that context. Since another poster read your response in way similar to the way I read it, it suggests that your response is the one at fault.

Sivaraman Dhamodharan

Thanks Crow. That is constant piece of memory.

Programming Article