This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: C++'s 'export' Keyword?
On Wed, Jan 28, 2004 at 05:35:07PM +0100, Gabriel Dos Reis wrote:
> Even if the the GCC SC choses to reject any implementation of export
> in GCC, we would still need points (1), (2) and (3) for various
> reasons.
The SC isn't going to reject implementation of "export"; RMS will drag
his heels a bit, but he's accepted GNAT's support of dumping and reloading
meta-information about Ada programs, even though it's the kind of
interface to GCC that he does not want, because it's required by the
language standard. Similarly, now that competing compilers are starting
to support "export", we're going to get increasing pressure.
At some point, someone will contribute a good design for "export", and the
SC will then spend several months persuading RMS to accept it.
I have some respect for RMS's point of view on this matter: he points out,
correctly, that GCC only supports C++ and Objective-C, as well as only has
backends for a number of processors, because there was no nice modular way
to attach a proprietary front end or back end to GCC. However, the other
side of the issue is that CS grad students interested in compilers are
increasingly working with something else other than GCC because what we
have now is not suitable as a research platform. Something will have to
give eventually.