[rrd-developers] regression testing?

[Tobias Oetiker]

Today Sebastian Harl wrote:

> Hi,
>
> On Thu, Oct 16, 2008 at 04:21:56PM +0100, Daniel.Pocock at barclayscapital.com wrote:
> > Are there any test scripts I should run before submitting patches?  At
> > the moment, my testing is limited to those things that I think might be
> > impacted by a change.
>
> I don't think there are any such test cases so far. I've already been
> thinking about introducing automatic testing mechanisms but I did not
> yet have the time to do so.
>
> I'm pretty sure, the basic mechanism can be copied from some other
> project. However, we need to think about what to test. So far, I've just
> come up with the following first ideas:
>
>  - use aggressive compiler flags, debugging memory management libs, and
>    similar (not sure though if that can reasonably be done automatically
>    but having the test cases available would still be a good thing)
>
>  - cover example use cases (e.g. starting with examples from the
>    documentation like the Beginner's Guide) and verify the results
>
>  - define the expected behavior for each tool and design test cases for
>    each specified feature (this will probably be a lot of work but, I
>    guess, it will provide the best test cases and might also further
>    improve the documentation)
>
> Any comments, suggestions, further ideas, and, especially, patches are
> mostly welcome ;-)

Way to go sebastian. As I just wrote in the other mail there is a
make test target in the perl bindings, but the scope is very
limited ... something more encompassing would be most welcome.

cheers
tobi


> Cheers,
> Sebastian
>
>

-- 
Tobi Oetiker, OETIKER+PARTNER AG, Aarweg 15 CH-4600 Olten, Switzerland
http://it.oetiker.ch tobi at oetiker.ch ++41 62 775 9902 / sb: -9900