Subject: Re: err(3) and error handling
To: Kenneth Stailey <kstailey@leidecker.gsfc.nasa.gov>
From: VaX#n8 <vax@ccwf.cc.utexas.edu>
List: current-users
Date: 02/09/1995 04:16:44
FWIW, I'm pretty sure a decent job could be done in user-land.
It'd take a little setup perhaps, but you could advertise all kinds of
nifty features (NLS, application-defined errmsgs, etc).

I was kind of interested in how people who have worked with large bodies
of code (i.e. NetBSD) might address such a problem.  Now I know - "last" :)
I can surely see the wisdom in evaluating the necessity for major changes
very carefully, though.  Certainly a re-vamp of the error codes and how
the system handles them, and updating all the source is no small gruntwork.

P.S.> One time, several years ago, I saw that "Not a typewriter" error message
and was convinced it was a joke :)

----Silly part follows----
How about just reducing the list to two error messages?

EINVAL - invalid argument to function, program, or method
         One of the parameters or arguments to the action you requested was
         not valid.
ESTATE - universe not in appropriate state
         The universe was not in a state where the action you requested
         could be (meaningfully) applied.  Please check power cords,
         connectors, phase of moon, fnord media, and try your call again.
-- 
VaX#n8 (vak-sa-nate) - n, CS senior++ and Unix junkie - vax@ccwf.cc.utexas.edu
23 Skidoo!                                            - PGP key on request