Single line pointer operations

I've been struggling with a homework problem in C++.

the problem is as follow:

Write one statement that performs each of the following tasks. Assume that num1 is a floating-point variable. Assume also that ar1[100] and ar2[100] are arrays of type char:

1. Declare ptr as a pointer to an object of type double
2. Assign the address of variable num1 to ptr.
3. Display the value of the object that is pointed to by ptr.
4. Assign the value of the object pointed to by ptr to variable num1.
5. Display the address of num1.
6. Copy the string stored in ar1 into ar2.

my problem is with number 2. the problem is that the pointer is a double but the variable num1 is a float.

my other problem is that i can't see how 6 could be done without the use of a for loop.

so far I have gotten this:

1. double *ptr = new double;
2. ptr = (double *) &num1;
3. cout << *ptr;
4. num1 = *ptr ;
5. cout << &num1;
6. //dont know

this gives me weird values if done but if i declare num1 as a double and change 2 to ptr = &num1 everything works fine.

Pl开发者_如何学编程ease help I have been on this problem for two days.

double and float are both floating-point types. So if your teacher said that num1 is a "floating-point type", you should be able to get away with using a double. If your teacher said to use float, then you're stuck.

6 is asking you to do something very specific. It will involve some googling, but it can be done as a 1-liner.

For the problem:

Assign the address of variable float num1 to double* ptr.

Are you sure that you're supposed to assign the address and not a value? To assign the address of a variable to a pointer of a completely different type is not meaningful.

You should also know that if you are using C++ as your tag says, you should prefer to use a C++-style cast (reinterpret_cast<type>(variable), etc.) to using C-style casts.

A double is a floating-point value, so ptr = &num1 should be ok. On the other hand, double *ptr = new double; not only declares a pointer to a double, but also initializes it. double* ptr is sufficient as declaration;

Casting a float* to a double* will likely not produce anything useful, because float and double probably have incompatible representations (particularly they are usually of different size).

See, double is a double-precision (binary64) floating point IEEE 754, while float is single-precision (binary32) floating point defined by the same standard.

So just declare num1 as double.

all seem right. except for number 1. I would just say double *ptr; It does not say to instanciate it, you are createing a memory leak.

6 is strcpy_s(arr2, arr1);

For 6:

string.h: strcpy(), strncpy() and strlcpy() - google them.