Re: Contents of <confdefs.h> for C++

[Akim Demaille]

>>>>> "Walter" == Walter E Brown <address@hidden> writes:

Walter> I had raised (rather courteously, I thought) some technical
Walter> issues in my several recent notes to this mailing list.

Why don't you expose your technical problem first.  Autoconf is not
built on theories, but facing problems.  Currently you exposed no
problems at all.  Being able to vote about C++ is a great thing,
really.  Now Autoconf is hear to cope with this language in real
situation.  What is the real situation you are facing?


Walter> At the moment, it is my opinion that autoconf is not
Walter> well-supporting the C++ community.

We all share this opinion.

Walter> I have previously offered rationale for my views, and have
Walter> even suggested alternatives for consideration.  I'd like
Walter> someone to make an authoritative decision on the need for
Walter> implicit, undocumented support for a "feature" that runs
Walter> contrary to established C++ philosophy in general, and that
Walter> has the potential for breaking (and has broken) user code.

This claim was never supported by actual code.  I second what has been
said here: Autoconf is not here to be beautiful, but to answer
concrete problems.  Each time someone asked you what your concrete
problem was, you answered in text instead of Autoconf code.


Walter> I regret that I will be away from my office for several weeks
Walter> while my wife recuperates from surgery (scheduled for early
Walter> tomorrow morning -- your thoughts and prayers on her behalf
Walter> would be welcomed), 

All the best!