Patrick Santucci wrote:
>
> I beg to differ... One moment while I get up on my soapbox...
[snip intereting suggestions about test data]
> I like to test programs with the *wrong* data just to see how robust
> they are: does it abort if the input file has the wrong record length or
> does it go merrily along assuming the data must still somehow be the
> correct data? Does it write exception data to another file or does it
> try to shove it into the database anyway? Things of that nature.
>
> While no program is totally bug-proof, IMO they should be able to handle
> not only the expected error conditions, but also what I would term
> "catastrophic" error conditions, where invalid data getting into your
> system could totally muck things up (like two fields being reversed,
> etc.).
>
> Just my $.02
> Patrick

Just a quick comment: Before worrying over if a program will misbehave, I
think one needs to test if it will load and run with unchallenging data.

I have to add that given my recent experiences with a certain commercial
Unix/Oracle application package, at the moment I'd be happier if at least
they would test that the *$#% thing will run with any data - or for that
matter just test that all programs will just load without errors.  If it were
my firm, I'd expect my lawyers weren't let me ship code link this.

--
Richard Gambrell
Database Administrator and Consultant to Computing Services
University of Tennessee at Chattanooga, Dept. 4454
113 Hunter Hall, 615 McCallie Ave. Chattanooga, TN 37403-2598
UTC e-mail: [log in to unmask]   phone: 423-755-4551
Home e-mail: [log in to unmask]