Alvaro Fernandez writes:
[re: exceptions during array operations]
> Maybe it should be left to the programmer? Return a code or something?...
Things aren't that trivial. The committees are not composed 100% of
morons like myself. Ok, I'm hopeless, but some of the people on them
are actually pretty smart. They tried to address these issues and
couldn't come up with an agreeable solution. That doesn't mean that
solutions don't exist. Of course, they do (even though I don't know
what they are at the moment). But I do assure you that the issues
aren't trivial.
How does one "return a code or something" from
A = B + C
(much less more complicated cases)? And how about if there are
several different errors generated in such an expression?
Automatic arrays and other simillar things that can happen in a
procedure call before you actually get to the executable code of
the procedure raise other tricky issues. Note, by the way, that
it is perfectly possible to get stack overflow when calling a
procedure with no arrays at all (its just not as likely).
I don't have good answers to all the issues handy. But just
dismissing everything as being a symptom of design by an incompetent
committee shows failure to comprehend the underlying issues.
You know, world hunger is a trivial problem also. Just feed them
or something. :-(
--
Richard Maine
[log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|