Re: [Ada] Bootstrapping mainline GNAT fails

[Alexandre Oliva <aoliva at redhat dot com>]

On Mar 19, 2002, dewar@gnat.com (Robert Dewar) wrote:

> However, dealing with arbitrary compilers that are broken (as in the
> case which caused this long thread) is another matter entirely.

I agree.  That's not the issue I'm talking about, even thought it was
such a thread that got me to express my opinion about my impressions
about the mindset related with GNU Ada.

> For example, there is a quite concious attempt not to use 3.14
> features yet, and we verify that this is the case.

It's the `yet' that bothers me.  See, this feels like a plan to
actively break things that currently work, at some possibly distant
point in the future.  If it works now, why not keep it working
forever?  I know, I know, you want to use features of new versions of
GNAT in the compiler itself.  But if the C front-end can manage to
build with earlier versions of GCC and non-GCC compilers, while
taking advantage of features in newer versions, couldn't the Ada
front-end take a similar approach?

Hmm...  Here's my total ignorance of Ada gets in the way: I wonder if
Ada has some feature that resembles a C preprocessor, that would allow
the auto-selection of chunks of code based on features available in
the bootstrap compiler.

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer