This morning I changed the way the logger reports pass/ fail to better
report failure when any test fails or entire suite fails to complete
(commit r9494). I also moved John's new TestBlocks to be the last test
to run, as it seemed to be leaving Chandler in a state that caused
problems with many other tests On windows at least this seems to work
as a temporary solution, I am watching tinderbox to see how the other
OS's fair.
In the longer term I would like to see the Functional tests be able to
be run in any order.
Dan
Aparna Kadakia wrote:
I talked to John about this briefly on Friday and it seems that this
error may be related to the state of the app carried over from the
execution of the previous tests.
I am planning to run the same tests with a new repo copied over for
each test to see if we can reproduce the same error.
Aparna
On Feb 13, 2006, at 7:24 AM, John Anderson wrote:
That's true, however, when I did that the test still fails, just in
a different way -- it hangs.
John
Andi Vajda wrote:
On Sun, 12 Feb 2006, John Anderson wrote:
Yes, the functional test are broken.
I spent most of Friday looking into this, but didn't find the bug.
It turned out to be very complicated. The test randomly fails in
different ways and some times doesn't fail at all. However I
learned a lot which we're following up on. If you want to hear all
the details, let me know.
I looked around in svn and it appears that rev 9464 introduced the
current behaviour. If you refresh to rev 9463, the usual 'show'
happens and finishes with a claim that 'all tests passed'.
Andi..
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Open Source Applications Foundation "Dev" mailing list
http://lists.osafoundation.org/mailman/listinfo/dev
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Open Source Applications Foundation "Dev" mailing list
http://lists.osafoundation.org/mailman/listinfo/dev
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Open Source Applications Foundation "Dev" mailing list
http://lists.osafoundation.org/mailman/listinfo/dev
