[Info-vax] Looking into C-include files on VMS

[Bob Eager]

On Sat, 31 Oct 2009 22:15:52 +0000, Joerg Schilling wrote:

> In article <7l3bsvF3a4r6dU2 at mid.individual.net>, Bob Eager 
> <rde42 at spamcop.net> wrote:
> 
>>Of course, that's what you have to do; think in the paradigm of the
>>current system. Each system has its own merits, and neither is "better"
>>than the other. VMS heavyweight processes are, however, a pain at times.
>>The problem with 'configure' is that it was designed to handle
>>portability problems across UNIX systems, and was never designed for
>>something as different as VMS. And many UNIX adherents think that
>>'configure' is a crock anyway.
>>
>>And yes, I've done a lot of programming on both systems; 26 years on VMS
>>and 33 on UNIX...
> 
> Then you should know that "configure" implements ideas (fine grained
> tests) that have been developed by the OpenSource movement in the late
> 1970 to the early 1980s.

And it's a mess. Time to throw it away and start again.

> The basic idea is to use few commands that may
> be easily ported by hand before.

I'm aware of that. I have actually used it.

> These commands are part of POSIX
> anyway, so any OS should have them since aprox. 20 years.

Any POSIX-compliant OS. Which is what I said...it's only intended for 
UNIX-like (POSIX compliant) systems.

> If you however follow the documentation from the FSF that comes with the
> "autoconf" project, you make a mistake and then you are indeed at the
> dark side of the force ;-)

'autoconf' is an integral part of the broken system. 

> This is why the Schily Makefilesystem does not use "autoconf" the way it
> is documented by the FSF but the way it makes sense.

OK, so you have a homebrew system that looks a bit like 'configure', and 
it *may* work. Thus it is not the original 'configure' system, and my 
assertion stands.

-- 
Use the BIG mirror service in the UK:
 http://www.mirrorservice.org