Re: [RFC] type safe trees

[Mark Mitchell <mark at codesourcery dot com>]

Gabriel Dos Reis wrote:

> You Wrote Nathan Sidwell
>  
>
> > Gabriel Dos Reis wrote:
> >    
> >
> > > So, we don't want C++, we find C too restrictive in the expressive
> > > power it gives us and we end up inventing a new language we
> > > have to learn?
> > >      
> > yes, because C++ is not a perfect match either
> >    
>
> I was not pretending C++ is a perfect match, but surely there is
> a good subset of C++ that enhances C to the point of having
> something better.  From the sketch of the new "language",
> I'm skeptical that it is a perfect match.
>  
Seriously, how about we just require that GCC be built with a C++ 
compiler that supports whatever features it is that we require?  At 
first, that would be classes, with single inheritance, and no member 
functions.  Nothing more -- no templates, no virtual functions, no 
exceptions, etc.  That's a subset that even Cfront handled 100% 
correctly, AFAIK.  We could add to the subset over time, but the goal 
would be to keep it small enough that we never had any trouble 
bootstrapping.

(We can gradually include things like pieces of the standard library.)

I think the bootstrapping argument is a red herring if you stay within 
this subset, which we could inforce with warning options to G++.  There 
are only a few host systems that people really use, and they all have 
precompiled versions of G++.

--
Mark Mitchell
CodeSourcery, LLC
(916) 791-8304
mark@codesourcery.com