Re: yet another C++ implementation bug to work around in Bison

[Hans Aberg]

On 6 Feb 2006, at 20:26, Paul Eggert wrote:

> Hans Aberg <address@hidden> writes:
>
> > I guess the GNU philosophy that anything beyond GCC is extra. :-)
>
> Well, I've been spending the last month or so trying to make sure that
> Bison 2.2 (when it comes out) will work with all sorts of bizarre
> non-GCC environments.  So your comment came as somewhat of an
> unpleasant surprise.  It appears that you haven't noticed those
> efforts.  Certainly the goal is for Bison to work with all sorts of
> weird compilers, just as it has for many years.

I guess to put something into my comment that wasn't intended. :-)

> > I guess I do not know the world of contemporary C++ compilers that
> > well.
>
> There's no shame in that; hardly anybody does.  (I certainly don't.)
> But we get bug reports, and we try to fix them.

I just have GCC 4.0.1, which suffices for me. :-)

Templates are tricky to implement, but if there is a method to do it,  
I thought compilers would have caught up by now. But on another  
compiler I used, I noticed linking errors, due to the fact that it is  
hard for the compiler to figure where to put the object code, if the  
template generating the code is in a header. A similar problem arises  
with "inline" functions in a header that cannot be inlined, I figure.

  Hans Aberg