...

template struct nontypeadapt;

template struct nontypeadapt { typedef Ret result_type; Ret operator() () { return func(); } };

then

typeid(boost::bind(nontypeadapt())) != typeid(boost::bind(nontypeadapt()))

However, the signature has to be given at the call site, it's repetitive to define all specializations of nontypeadapt, and it brings more forwarding problems of it's own. Any ideas?

I think you can only pass functions with external linkage as parameters to templates, which I think it makes it unusable...

I believe that's true, yes, but it's usable nonetheless. The assembler generated by vc71 for a bind with nontypeadapt is better, too. Bruno

Simon Buchan writes:

Bruno Martínez wrote:

...

Hi.

When using pointers to functions with bind, the particular function doesn't play a part in the return value's type of bind. That is

void func1(); void func2();

typeid(boost::bind(&func1)) == typeid(boost::bind(&func1)) // this is true

I need the types to be different to store bind objects with the serialization library. The C++ Templates book proposes an adaptor that takes the pointer to function as a template parameter. However, this adaptor is particular to a given signature, and I don't know how to rewrite it more generally. Is this possible? Can I more easily make boost::bind do what I want?

Bruno You may be making more work for yourself than you need to, why do you need to serialize bind objects? I haven't used the serialization lib, but a quick look-through doesn't really bring any times to mind where I would want to serialize a boost::bind functor!

I would like to do the same thing - boost::bind can be used to bind an instance of an object, a member function pointer _and_ any parameters required by that function. This leads to a really useful implementation of the command pattern - you can build a list of these functors which constitute a composite command, and to serialise the whole lot - or part - would be invaluable in all sorts of situations including tracking how much of the composite command had actually been implemented. Maybe I should generalise and boostify my implementation. Any interest? S

Bruno Martínez writes:

On Thu, 15 Sep 2005 10:47:41 -0300, Simon Carter wrote:

...

I would like to do the same thing - boost::bind can be used to bind an instance of an object, a member function pointer _and_ any parameters required by that function.

This leads to a really useful implementation of the command pattern - you can build a list of these functors which constitute a composite command, and to serialise the whole lot - or part - would be invaluable in all sorts of situations including tracking how much of the composite command had actually been implemented.

Maybe I should generalise and boostify my implementation. Any interest?

Yes. Could you explain it's principles? I've been trying to solve the same problem for some time now.

Sorry, perhaps I wasn't clear enough - I haven't solved the general problem of the serialisation of bind or function objects using boost at all - though I would like to do the same thing. The implementation I have is too specialised on persisting just the bits I need, and uses a our own persistence mechanism. I'm afraid it wouldn't be much of a starting point for a general solution, though I too would be very keen to see one.