Dallman Ross wrote:
> Speaking of TRAP, exit codes, etc., if I set an exit code
> of, say, 1 for a certain condition, would that exit code
> not bounce mail (under anybody's mail system)? I do not
> wish to bounce the message in the scenario I have in mind,
> but I do wish to set exit code, either via TRAP, or via
> EXITCODE in procmail, to non-zero in a certain case.
I wonder what would happen if the exit code is less than 64.
In any case, the question is incomplete: what are you trying
to do with the non-zero exit code? Just log it? Report it
back to the shell when you're invoking procmail interactively?
Can you set it in a clone procmail so that it goes back only
to the parent procmail and not to the MTA?
I could probably do the latter, but it's way more trouble
than it's worth for what I have in mind. Let me explain.
Here's an includerc that calls another includerc. Now
I test to see if the called includerc actually ran.
If it didn't, I bail out of what I was going to do.
That's all simple enough. But I don't really want to quit
with EXITCODE = 0, the default. I am sending a message
to the log that we were fubar; and I think an exit code
that is non-zero is an appropriate thing to have showing the
same thing -- just for canonical programming style if for
no other reason. But I don't want the MTA to do anything
untoward (like bouncing the message) -- the message hasn't
been harmed, and it can just got do $DEFAULT now or wherever
it was going to go after we bail from the failed stab at
something.
> (And what about negative numbers?)
I think exit codes are unsigned eight-bit values.
Hmm, okay.
_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail