Hi Jason,

did you use a refactoring for this particular method before the
problem appeared? It seems to me that the method was not well compiled
and a bug like this existed some month ago. You might want to try
using a newer image or update the RefactoringService package.

Bye

2007/10/12, Jason Johnson <jason.johnson.081 at gmail.com>:
> Hi all,
>
> I am running Squeak web dev image: web07.08.1.  I have a test that
> uses a class side collection variable, where I check upon entry to the
> test method that the collection is empty.  Strangely enough, this
> assertion failed with one of my tests (it shouldn't as all tests end
> by clearing the variable).  This particular test did have some code in
> it that would fail, but it was a ways after the size check.  When I
> ran it through the debugger it worked to the real fail point, where I
> fixed it.  After that the test worked.
>
> So, is it possible that the tester in this image is restarting tests
> that fail for some reason?  The behavior is as if the test ran to the
> fail point, then restarted the method from the beginning.  And the
> first time this happened, I think it did this twice before failing, as
> my collection that should have been empty had 4 elements.  Twice what
> that test would enter.
>
> Thanks,
> Jason
>
>


-- 
Damien Cassou