JOAQUIN LOPEZ MU?Z wrote:
The idea is that you can alter only some part of the hashing scheme and benefit of the preexisting definitions: [...]
I think I see your point. You want something like this: template<class F> size_t hash_value( int x, F const & ) { return x; } template<class F, class T> size_t hash_value( vector<T> const & v, F const & f ) { return for_each( v.begin(), v.end(), f ).value(); } which, when combined with a suitable combiner 'f': class F { private: // state public: template<class T> void operator()( T const & t ) { // combine state with hash_value( t, F() ); } size_t value() const { // collapse state into a size_t and return it } }; will automatically propagate it downwards if one invokes hash_value on vector<vector<vector<int> > >. Interesting, but overkill, IMO. ;-)