From: Beman Dawes (bdawes_at_[hidden])
Date: 2001-02-08 20:01:10
At 10:17 PM 2/8/2001 +0100, Jens Maurer wrote:
>Personally, I believe that overloading in std should be ok, except that
>the set of restrictions must be worked out carefully. However, at least
>the simple void std::abs(const my_class<T>&) should pose no problem
>at all regarding potential ambiguities.
>Boost could be a test ground for such a resolution.
Maybe you should post such a proposal on both boost and the committee's
library reflector. I don't think very many committee members want to
change the core language rules, but AFAIK no one has stepped up to the
plate with proposed wording to allow user overloads in namespace std.
While normally boost doesn't deliberately diverge from the standard, it
seems to me to be perfectly natural for boost to act as a test ground for
such a resolution. And I'm sure the LWG would look more favorably on a
proposed resolution that had been tested in practice.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk