Re: Best practices

Niall Litchfield wrote:

> "Daniel Morgan" <dmorgan_at_exesolutions.com> wrote in message
> news:3D246FC0.EF6637A3_at_exesolutions.com...
> > The potential error was in fact two fold. One that it would not handle
> NULL. But
> > that was an issue QA could catch is the spec said an account type was
> mandatory
> > and this was enforced with NOT NULL and testable. What was not testable
> was that
> > the bank might, and I emphasize "might" add other account types later.
>
> I'm not entirely sure how that is different from my suggestion that QA put
> invalid values in the test data. though a list of valid account types ought
> to be testable with a check constraint. I confess i'd probably miss it on
> code review but spot it when one of my colleagues puts rubbish values
> through in sql*plus.
>
> > It is not testable because since those account types didn't exist at the
> time of
> > the testing every single test passed with flying colors.
>
> that is , as i suspect you know, one of the issues with testing. You don't
> just test to see does the ap work as designed you also test to see what
> happens when the app is used *not* as designed. my 99 hours of work a day eg
> was supposed to illustrate this. only consultants can record this much time
> <g>.
>
> > Amazingly enough, I recieved an email from an employee there yesterday
> that saw
> > this thread and said she found a similar situation in some code, again,
> just a
> > few weeks ago. Just about any IF statement that can't handle ELSE is a
> potential
> > disaster waiting to happen. And the place to catch it is in code reviews
> ... not
> > in production.
>
> absolutely. any thoughts on my record a value for each of the 48 half hours
> or 24 hours in the day?
>
> --
> Niall Litchfield
> Oracle DBA
> Audit Commission UK

The reference to 99 hours and 48 half-hours didn't make it here somehow.

You are correct that complete testing would have caught it. But that kind of testing never takes place. Testing, at best, uses data from upstream systems and feeds to downstream systems. No one seems to have the luxury of doing it correctly though they have plenty of time to clean up afterwards.

Still, I prefer code reviews led by someone that is senior enough to know what to look for and has enough civility to not turn it into a testosterone contest.

Daniel Morgan Received on Thu Jul 04 2002 - 15:10:02 CDT