Re: [AMBER] picky nab compiler: is this really necessary?

From: M. L. Dodson <activesitedynamics.comcast.net>
Date: Wed, 16 Jun 2010 12:54:37 -0500

On Jun 16, 2010, at 11:57 AM, case wrote:

> On Tue, Jun 15, 2010, M. L. Dodson wrote:
>
>> nab -c -O3 driver.nab -o driver.o && /bin/rm -f driver.c
>> nab: only 1 of -c or -o can be used
>
> It looks to me like you should be able to just comment out lines 100-104
> in nab.c (which produce the error message above). This assumes that your
> C compiler accepts both -c and -o; some (old??) compilers did not. But
> basically it looks like this limitation serves no real purpose.
>
> ...thanks for your comments...dac
>


Thanks for the explanation. Lest I come off as just a pest, I will say that
there are good reasons (which would be OT here) for compiling software systems
from read-only file system mounts, and the FreeBSD (all BSD?) people have made a
fetish of enforcing this kind of construct in their system software. I just
tested, and the capability to use both -c and -o goes back to at least gcc-3.4.6
as implemented on FreeBSD-6.3. I've been using this OS since the compiler was
gcc-2.x (maybe 1.x?, FreeBSD-2.0, ca. 1992), so unless they were less strict
back in the day, I assume it worked there as well. I can't speak to other
flavors of Unix or Microsoft compilers since I have never tested for the
capability with them.

>From the point of view of software design esthetics, the two functionalities
implied by -c and -o, if not orthogonal, are nearly so, implying logically they
should not interfere. IMO, of course.

Thanks again,
Bud Dodson
-- 
M. L. Dodson
Business email: activesitedynamics-at-comcast-dot-net
Personal email: mldodson-at-comcast-dot-net
Phone: eight_three_two-five_63-386_one
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Jun 16 2010 - 11:00:03 PDT
Custom Search