MS DevStudio 97, VC++ 5

This elementary de{*filter*} operation has *got* to be in here somewhere, but
I can't seem to find it. Can anyone tell me the magic
command/menu/keystroke?

I have a pointer (lets say to type double) the de{*filter*} correctly shows me
the value pointed to, but the de{*filter*} is unaware that it is merely the
first of many values in an array! (I'm viewing the guts of a container
type class)

How do I view the array of doubles instead of just the first one?
I can't find anything on this in the online help files.

--
Tim Gogolin
Software Engineer, Stat-Ease, Inc.

There are two work arounds, the first in the de{*filter*}, the second in
code :

double_var[0]
double_var[1]  up to
double_var[n]  show you all you want;

In code :

#define NBR_OF_DOUBLES 17

int func(double * ptr)          // actually an array
{
        double arr[NBR_OF_DOUBLES];

        memcpy(arr, ptr, sizeof(arr));

Since a pointer is syntactically the same as an array, the de{*filter*}
can't see the difference. So you have to show it, don't you?

For builtin types, it's easy.  Just open up the memory window and
point to that address.  You can switch it to view memory as doubles,
either in the context menu or in Tools.Options.Debug

For user defined types, it's a little tougher.  One option is to
create an actual array:

void func (Foo foo*, unigned len) {

#ifdef DEBUG
        Foo foo_copy[1000];
        memcpy (foo_copy, foo, sizeof (Foo, len);
#endif

Sorry, they're not the same.  Try:

// file1.cpp

extern int *p;

int main () {
        p = 3;

        return 0;

int p[100];

It won't work.

--

Use the same syntax as you would in a program.

e.g. double *pDouble;

then in the de{*filter*} watch window
        *(pDouble + n) where n is the element you want.

I guess I'm just spoiled by Metrowerks de{*filter*} (Mac version, I can't
speak for their PC offerings), which offers a "Open Array Window" menu
command when an array or a pointer is the current selection.

--
Tim Gogolin
Software Engineer, Stat-Ease, Inc.

Most hardware does not like to interprete any pointer as one to a
double; we must be aware of that, don't we???

Wim