Re: [Dev-C++] address pointers

[Per W. <pw...@ia...>]

Hmmm.

I assume your example was ment to be:

int x,y;
int *p = &x; /* Don't forget the address operator here... */
p += 1;

/Per W

On Sat, 31 Jan 2004, Michael C Urban wrote:

> On Sat, 31 Jan 2004, Brian Gordon wrote:
>
> > I'm new to C++ and in my C++ book, I am learning about pointers. I have
> > a question: can you store an address to some kind of "omnitype" pointer?
> > I am thinking about doing something like this:
>
> Although in my last message I showed you how to do what you are asking, I
> would also stress that directly manipulating pointer addresses is usually
> quite risky unless you are doing some kind of very low level programming
> (like programming a microcontroller). Pointer address manipulation is only
> guranteed to behave well if you are dealing with an array, where you know
> the the elements of the array are stored contigiously in memory. The
> following code sample is a disaster waiting to happen:
>
> int x, y;
> int *p = x;
> p += 1;
>
> In this case, I have increased the address of *p by one integer offset. In
> other words, I have increased the address that *p points to by 32 bits. So
> *p no longer points at x. But does it point at y? Probably not, because
> there is no gurantee that x and y are stroed contigiously in memory, even
> though they were declared one right after the other. So in the above
> sample, we cannot predict what the results will be since *p now points to
> an address with unpredictable contents. (More then likely, *p now points
> to garbage).
>
> Hope that clears things up, and hope I wasn't just telling you stuff you
> already know. But you said you were new to C++, and it looked like what
> you were doing was probably unsafe.
>
> Mike
>
>
> -------------------------------------------------------
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and Integration
> See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> Dev-cpp-users mailing list
> Dev...@li...
> TO UNSUBSCRIBE: http://www23.brinkster.com/noicys/devcpp/ub.htm
> https://lists.sourceforge.net/lists/listinfo/dev-cpp-users
>