From: John Torjo (john.lists_at_[hidden])
Date: 2003-07-22 13:38:07
> John Torjo wrote:
> > Yes, I remain unconvinced ;-)
> > This is because some compilers do not provide such a FUNCTION facility.
> > VC6 is one of them. What should I do for it?
> It sounds like you are asking for the wrong macro!
> You are trying to support a compiler that is outdated and long since
> replaced (but remains in widespread use) A FUNCTION_NAME patch is
> unlikely to become available.
> Rather than assuming BOOST_FUNCTION_NAME is not available unless
> implemented though, you really want a macro that says 'this compiler is
> broken and will never be fixed', BOOST_NO_FUNCTION_NAME or similar. So
> long as we are strict on allowing compilers into the list, I think all
> interests might be served?
Yes, it looks ok.
But I guess we're on the same side ;-)
This is what we wanted with BOOST_HAS_CURRENT_FUNCTION : just to tell us if
the current compiler has a FUNCTION_NAME facility.
(so, it could be renamed: BOOST_HAS_FUNCTION_NAME)
If we find that a current compiler has a FUNCTION_NAME facility, we (the
maintainer, of course) update the current_function.hpp header.
Where am I wrong?
> Unsubscribe & other changes:
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk