Strategies for Fixing Expansion and Truncation Errors

Here are some strategies for handling type expansion and truncation errors.

• Use a pointer-sized (64-bit) variable to avoid expansion or truncation during assignment.

• When you are sure it is safe to do so, use a hard cast to truncate a 64-bit value, such as size_t, to 32 bits, as shown below:

std::vector<int> myVec;
long nVecSize = (long)myVec.size(); // no warning

• When you are sure it is safe to do so, replace a 64-bit incoming value with a 32-bit value, as shown below:

extern long myGetSize(const std::vector<int> & ); // applies the              // truncation hard cast internally
std::vector<int> myVec;
long nVecSize = myGetSize(myVec );

• Temporarily disable the compiler warning and then re-enable it, as shown below:

std::vector<int> myVec;
// we know size is < 4G, so is okay to ignore warning
#pragma warning (push)
#pragma warning (disable: 4267)
long nVecSize = myGetSize(myVec);
#pragma warning (pop)