From: David Abrahams (abrahams_at_[hidden])
Date: 2000-10-14 19:09:27
----- Original Message -----
From: "Jens Maurer" <Jens.Maurer_at_[hidden]>
> I've run our regression tests through it. OpenC++ translates C++ to
> C++, so it still requires a usual C++ compiler; I tested with gcc.
> Features missing:
> - explicit instantiation
> - T::template f<int>() (i.e. member template access)
> - using-declarations (in some situations)
> - typename in dependent types in formal template value
> - typeid
> - operators.hpp gets garbled so that gcc is no longer able to
> compile it.
> I'm not impressed.
> Should this be part of the "official" regression test suite, just
> for fun?
Maybe instead we should send our regression suite to the OpenC++ people so
they can get their act together? That sounds like it would be more useful in
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk