On Sat, 08 Jan 2005 01:10:03 +1100, Andrew Cowie
<an...@op...> wrote:
> On Fri, 2005-07-01 at 08:50 -0500, Jeff Morgan wrote:
> > And now on to the question. During our next development cycle we hope to
> > introduce a testing framework for the java-gnome bindings. The question is
> > "should this testing framework be based on JUnit or Mauve?". I am interested
> > in your experience with both of these tools.
>
> [cut and paste from #java-gnome]
>
> Unit testing GUI things is hard. How do you prove that a GUI element
> actually popped up? You can fire events, but then it's not actually the
> mouse click that causes the event, it's some manual trigger which
> doesn't exercise the full code path.
Thomas Fitzsimmons wrote some code for the classpath project
that provides a wrapper around the XTest library. I am sure this
code could be used for java-gnome. I am also convinced that
much of the bindings can be tested without even displaying a
window.
> [I've been grappling with this as I try to figure out how to test the
> apps I'm writing. However we approach this, I'd like the solution to be
> something VERY usable by people using the j-g bindings to write apps, as
> opposed to just the people hacking on the bindings themselves]
>
> FWIW, my vote would be JUnit since a) eclipse has extensive support for
> it b) it's widely used in the Java community.
I just wrote a simple JUnit test for the ComboBox class. My goal was to
see how easy it would be to write JUnit based tests for Java-GNOME.
I have attached the source here. Please take a look and let me know
what you think.
> Andrew Frederick Cowie
>
> OPERATIONAL DYNAMICS
> Operations Consultants and Infrastructure Engineers
> http://www.operationaldynamics.com/
>
> Sydney: +61 2 9977 6866
> New York: +1 646 472 5054
> Toronto: +1 416 848 6072
> London: +44 207 1019201
--
Jeffrey Morgan
"The highest reward for a man's toil is not
what he gets for it, but what he becomes by it"
- Jon Ruskin
|
