[Qgis-developer] Rethinking the testing and release procedure of QGIS

[Martin Dobias]

Hi Marco

On Thu, Jul 7, 2011 at 5:23 PM, Marco Hugentobler
<marco.hugentobler at sourcepole.ch> wrote:
> Hi all
>
> Being in favor of unit tests, we cannot hope to solve with them all the
> stability problems, as some mails here suggest. It is just one small piece. In
> fact, thinking about bugfixes in the past, e.g. for the 1.7 release, most of
> them occured as a complex combination of several factors and user interaction,
> very hard to detect with unit tests.

I believe that many stability problems are caused due to various
unhandled corner cases in the core library. Bugs in GUI are usually
less fatal.


> Another point that needs to be considered is that the unit testing code needs
> to be maintained. Adding a test for every little change, even bugfix, would
> create a huge amount of testing code and sample datasets.
>
> What about alternatively creating a unit test base that is limited in size but
> well maintained, e.g. a good coverage of the core classes?

I don't think we need to aim for 100% coverage by unit tests. For me
the most important elements for unit testing are classes which
facilitate data access (map layers, features, geometries, providers
etc) - to make sure that we do not get regressions there.

Martin