asked on

size() of vector gives type conversion warnings

Hi,

I recently inherited this library (the source code) which uses STL vectors. I'm running it in VC++ v7.0.9466 and .NET framework 1.0.3705, and everytime i compile, i get a whole bunch of warning messages about "conversion from 'size_t' to 'blah', possible loss of data."

For example, the code below will produce 2 such warnings.

#include <vector>
void main() {
      std::vector<unsigned int> temp(5);

      int foo = temp.size();
      unsigned bar = temp.size();
}

I understand that the size() method returns a type of size_t, which explains the warning messages. But is there a way to make VC++ detect that for example it is actually an unsigned and not produce unnecessary warnings? Even though the warnings are harmless, I just don't like getting tons of warning messages everytime I compile.

Thanks.

Jaime Olivares

size_t is in fact an unsigned integer, so you can use:

unsigned int foo = temp.size();

But, why not to use this instead?

size_t foo = temp.size();

Anyway, to avoid an specific warning message you can use a pragma directive in the top of your files:

#pragma warning( disable : 9999 )

where 9999 is a warning code.

ASKER CERTIFIED SOLUTION

drichards

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

Selcon

ASKER

is that the best way to deal with these warnings?

so in all projects that use STL vector, there will be static_cast<> all over the place, especially in loops across elements in a vector, etc.? or there will be size_t all over the place i suppose.

drichards

Yes, without disabling the warning, those are your choices.

I try to use size_t where I can and then use static_cast to handle cases where the target value is already a smaller integer type.