Re: [Boost-users] enable_if and templated operator() of member of class template

2 Mar 2012


      On Fri, 2 Mar 2012, MM wrote:
...
...
-----Original Message-----
From: boost-users-bounces@lists.boost.org [mailto:boost-users-
bounces@lists.boost.org] On Behalf Of Jeremiah Willcock
Sent: 01 March 2012 16:48
To: boost-users@lists.boost.org
Subject: Re: [Boost-users] enable_if and templated operator() of member
of class template
On Thu, 1 Mar 2012, MM wrote:
The main issue here is that the condition in enable_if does not depend
on
the template parameter to operator(); see the version in
http://codepad.org/6AORo7tU, in which I've used boost::enable_if and
added
some tests.  One fix is to make the enable_if condition dependent, as
in
http://codepad.org/916Tl6W8 (there were a number of other issues in the
code; note that the is_same condition needs to
be
in the enable_if since the body of the function will not always compile
when that is false).  Please see if that last paste works for you.
-- Jeremiah Willcock
Thank you, http://codepad.org/916Tl6W8 compiled. Using std::enable_if as
well. (msvc10 uses their TR1 impl).
I need an extra check for the 2nd operator() of the visitor, that
variant_iter_t is actually identical to iter_t::Iterator embedded type.
Basically, it treats the case where iter_t is a reverse_iterator of the
iterators in my variant, and we need to check that we're applying visitation
on the proper iterator.
So   line 43: is_same should be 
the problem is that this operator is instantiated even for the case where
iter_t is a const_iterator and not a const_reverse_iterator and compilation
fails, because const_iterator has no embedded type Iterator.
How to fix?
Try this version instead: http://codepad.org/uA7oflvk.  It uses a helper 
template to do the dispatching among the overloads.

-- Jeremiah Willcock