I know there are plenty of questions about the argument, but I still don't understand some basic stuff about rvalues.

Supposing I have a function of this kind:

/* 1 */ void func( std::string s ) { /* do something with s */ }

which I use in this way:

int main()
{
  func( "a string" );
}

In a case like that, in which I pass this string to the function func wouldn't be better to always define this kind of function in this way:

/* 2 */ void func( std::string&& s ) { /* do something with s */ }

My question is: wouldn't be always better to use /* 2 */ instead of /* 1 */ in cases like the one of this example? And if not, why and where should I use the one or the other? The fact is that, if I understood well, in /* 2 */ the parameter is moved and will never be used outside of the function scope, right?

CodePudding user response：

In your specific example where a temporary string is created at call time, both expressions are equivalent.

Indeed, you could not keep both and use overload resolution, because they capture the same type of object (again, only in your example). They both can capture the temporary string and have it available in parameter s.