🚀 go-pugleaf

Philip Guenther (guenther@gac.edu) wrote:
: 
: In article <1995Sep11.221423.6651@forte.com> mikes@forte.com (Mike Schilling) writes:
: > Still, writing largely portable programs which 
: > use OS-specific features where possible and appropriate would be far eaiser
: > if this kind of thing were encouraged and where possible mandated.
: 
: This statement has left me almost speechless.  I presume that you
: think that your sys_exit() function is an example of such a
: feature. Can you please provide an example of *portable* usage of
: your sys_exit()?  You'd like to encourage the usage of
: implementation specific behaviour?  What's the point of the
: standard then?
: 
Your response leaves me almost speechless :-)

Of course sys_exit couldn't be called portably; its purpose in life
is make accessible the non-portable part of what 'exit' does.  (I
don't claim it's the best possible implemntation; I later suggested
another, as did Jeremy.)  The problem with 'exit' is that it isn't
either wholly portable, like strcmp, or wholly system-specific, like
system; it mixes the two inextricably. (At least, there's no portable
way to separate them.)

I don't know what you mean by 'encourage' the usage of implementation-
specific behaviour.  If an application I'm writing will benefit
significantly from OS-specific features, I'll use them, yes.  My assumption
is that customers will be unimpressed by hearing that a feature they
care about couldn't be implemented in portable C.  

And the purpose of the standard IMHO is to make my job easier.  Whatever
it can do in that direction is A GOOD THING.

Mike