[boost] forwarding function problem (was Re: Ongoing review of printf-like formatting)

From: Samuel Krempp (krempp_at_[hidden])
Date: 2002-01-26 11:31:24

On Fri, 2002-01-25 at 19:50, Peter Dimov wrote:
> > What is known for sure, is that gcc and comeau both accepts the
> > following :
> 
> Yes, you are right. I remember this being reported as ambiguous by the EDG
> front end but this must have been an older version.
> 
> In any event, the active issue still has no resolution.
> 
> http://std.dkuug.dk/jtc1/sc22/wg21/docs/cwg_active.html#214

Is it safe to assume that when the issue will be closed, the 
T& / T const& overloads will be uncontestably accepted by the standard,
or is it likely to be the opposite ?

If those overloads will eventually be fully legitimated, we can expect
more and more compilers to disambiguate those calls in the future

If not, I should remove those overloads from Boost.format, even if it
works right now for (at least) gcc and comeau.

I've come to read bind's documentation, and realised this was a well
know problem. So I guess there must be some established policy on
whether or not rely on those tricky overloads as a solution to the
'forwarding function problem'.

-- 
Samuel