On Fri, Jan 02, 2004, Jewett, Jim J wrote:
> For the moment, lets assume it was a sometimes problem, and his
> fix just simplified the code, which seems to have helped. 

I don't think programming by coincidence is a good idea.

> I'm not quite certain how to test for any of the three.  

That doesn't mean you can't test for the bugs, does it? I don't say
we should test *everything* in the code base (that's impossible); I
do think we should test things that we are most worried about going
wrong (i.e. make the test risk driven) and if we are about to fix a
bug then I would definitely say something went wrong. If we have a
test case for every bug then we will only find the bug once.

No one has said that it is easy to write test cases, but it's better
to have a few incomplete tests than no tests at all...

> I think a "Can we follow links embedded in tables" 
> test would make more sense than anything specific
> to Chris' first bug.

If there had been such a test then the bug would have been found
already during the screen redraw development, wouldn't it? That is,
there would be no reason to write a specific test case for the bug,
since it had been found (and fixed) before the new code was added
to cvs.

/Mike

_______________________________________________
plucker-dev mailing list
[EMAIL PROTECTED]
http://lists.rubberchicken.org/mailman/listinfo/plucker-dev

Reply via email to