Re: autoconf and STL

[David Boreham]

Bob Friesenhahn wrote:

> Rather than assuming that anything that could possible be broken needs 
> to be tested, I suggest that you start off with assuming that features 
> defined by the C++ standard actually work and only implement 
> work-arounds when necessary.  In most cases, compilation problems can 
> be resolved with very minor (and portable) fixes to the code.

Thanks, your words made me realize that I haven't done
any serious C++ coding in 9 years and a lot has happened
in that time* !

I realized that one of the problems I'm having is due to the
code I'm working with using deprecated C++ library classes
such as strstream, and non-standard extensions such as hash_map.
If I can get those things under control then I suspect it'll compile
with most modern compilers with minimal aucofonfing.

Thanks again for putting me on the right track.