<P>(I appologize for the two awkward replies I had sent before. It seems that my news client and myself have some issues to resolve...)</P>
<P><<A href="mailto:scleary@jerviswebb.com">scleary@jerviswebb.com</A>> wrote:</P>
<P>[snip]<BR>><BR>> 2) The Visitor concept brings a couple questions to mind:<BR>> a) Why allow modification (termed "persistency" in the docs)? No other<BR>> FunctionObject abstraction does, and modern C++ users are well-acquainted<BR>> with and used to the associated restrictions on FunctionObjects.</P>
<P>Indeed we had some discussions about this issue. The final argument was that<BR>the usage scenarios are different for a visitor and for (traditional)<BR>FunctionObject: In the STL context a FunctionObject is needed when one action<BR>should be repeatedly applied over a single sequence of objects. This task is carried out by a *single* invocation of the appropriate algorithm. A Visitor, on the other<BR>hand, is likely to be applied over several variant instances, which are not<BR>necessarily stored inside a container. Hence, the code will (likely) have<BR>several apply_visitor() calls, where a different variant object is passed to<BR>each one. If the visitor was passed by value, we would end up with something<BR>like this:</P>
<P>struct my_visitor { .... };</P>
<P>typedef variant<A,B,C> variant_type;<BR>void f(variant_type x, variant_type y, variant_type z)<BR>{<BR> my_visitor vis1 = apply_visitor(my_visitor(), x);<BR> my_visitor vis2 = apply_visitor(vis1, y);<BR> my_visitor vis3 = apply_visitor(vis2, z);</P>
<P> // do something with vis3 ....</P>
<P>}</P>
<P>I feel that this code is overly verbose and counter-intuitive.</P>
<P>[Two other points - in favor of the pass-by-reference method - are the<BR>ability to use polymorphic visitors without slicing them, and the reduced<BR>overhead, since copy-construction is avoided. However, these points are less<BR>substantial than the first argument, since you can use a simple wrapper<BR>class to work-around the problems.]</P>
<P><BR>> b) Do Boost.Lambda expressions qualify as Visitors? I am not familiar<BR>>enough with the implementation of either library to be able to tell...<BR>></P>
<P>A valid visitor must expose a nested type, result_type, which specifies its<BR>return type. I believe it should be possible (although I did not try it yet)<BR>to wrap a lambda expression inside a proxy class which will define<BR>result_type as void.</P>
<P>-Itay</P>
<P> </P>