Rodrigo de Salvo Braz writes:

On Mon, 22 Nov 2004, David Abrahams wrote:

...

Because often the result type depends on the argument type that is actually passed:

(_1 + 3)(4)

vs.

(_1 + 3)(static_cast("hello"))

But isn't it the case that any instantiated functor object will have a well-defined result_type? In your examples it seems '(_1 + 3)' would be instantiated to different types.

The expression _1 + 3 is a function object by itself, and that's what you pass to the algorithm.

Anyway, I guess a better solution to my particular problem would be for make_transform_iterator to be able to receive a result type and make an adaptable functor out of any UnaryFunction, right?

Yes. Patches welcome. -- Dave Abrahams Boost Consulting http://www.boost-consulting.com

Frank Birbacher writes:

Hi!

David Abrahams wrote:

...

Because often the result type depends on the argument type that is actually passed:

But he is binding a pointer-to-member-variable which definately has a result type (which is the type of the member).

For the reasons that they didn't make that particular result into an adaptable unary function, you have to consult the Boost.Lambda authors. -- Dave Abrahams Boost Consulting http://www.boost-consulting.com