Help with SUnit

Ken Causey ken at
Thu Jan 22 21:56:08 UTC 2004

Let me point out.  I started working with Mike on #squeak and was seeing
very different behavior than he is.  I was working in my development
image which has some other packages loaded into it:

ANSI Compatibility (r1)
Debugger - stub method creation (1.05)
HTTPClient (19)
MCInstaller (6)
Monticello (69)
PackageInfo (1.30)
Refactoring Browser and related tools (1.03)
SARInstaller for 3.6 (21)
SUnit-TestRunner (2)
SqueakMap2 base (0.91)
SqueakMap2 loader (0.92)
VersionNumber (1.0)
Whisker++ (++)

I then tried a basic 3.7a-5657 image like Mike and see the exact same
behavior he does.  Loading SUnit-Test Runner does not change it.

There is a difference in usage between the image that has all the
packages above (which I will call my 'dev' image) and the image without
them ('basic' image).  In my dev image I have the behavior I am used to
seeing that once you get into a debugger for a failed test you have to
step into a method or two to get to the point where you are looking at
the failing test method.  In that case (same exact test class) the
stream instVar correctly has a WriteStream.  In the basic image if I do
exactly the same thing, the debugger immediately pops up on the test
method, but at Mike points out, the instance variables are not properly


On Thu, 2004-01-22 at 15:41, Ned Konz wrote:
> Ah, but it doesn't finish. Instead the assert raises an exception, which kicks 
> the tearDown method.
> See if it works the way you want if you take out the assignment to stream in 
> your tearDown method.
> My own feeling on this is that probably the tearDown should be moved inside 
> the openDebugger... method along with performTest so you can see your context 
> better and you won't get tearDown called when an exception is raised.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url :

More information about the Squeak-dev mailing list