EXIT

[Fairlight]

On Fri, Mar 06, 2015 at 06:16:38PM -0500, Kenneth Brody thus spoke:
> On 3/6/2015 5:02 PM, Fairlight wrote:
> >Not unless system() has a return code, and probably not even then, since a
> >shell sits between the two instances.
> 
> That's the purpose of the SYSTEM() function, as opposed to the
> SYSTEM command.  (If it didn't return the exit value, what would be
> its purpose?)

Was SYSTEM reworked at some point?  Bear in mind that 95%+ of my clients
are still using 5.0.x.  If there's a 5.6+ -ism, then I won't necessarily
know about it...or remember it.  I know about nested calls.  I did read the
5.8 release notes.

That license manager is killing you folks.  Nobody I know who isn't already
upgraded wants to be chained to it, no matter what enticements you put in
the main program.

Anyway, AFAIK, SYSTEM as an internal to fP was a wrapper around system(2).
If you rewrote it to emulate that behaviour, then I can see how you finally
get the actual exit code of the end target.  Intermediate shells aren't
necessarily going to give you their last child's exit code, so I wouldn't
think you could if you still wrapped system(2).

When, if ever, was SYSTEM redone?

mark->
-- 
Audio panton, cogito singularis.