On 3/2/07, Kelly Campbell <[EMAIL PROTECTED]> wrote:
One problem with Rob's method of initialization is you have to
re-initialize the broker for each unit test which I'd rather not do
for performance reasons.

The problem with the recommended approach James mentioned in the docs
(turning persistence off), is what about when you specifically want to
test the persistence functionality and how your application uses it?

The only difference enabling/disabling persistence makes is if you
reboot the broker during the test, so it shouldn't make much of a
difference when testing your application code.  i.e. the purpose of
the test is to test your application code - and during the test the
broker is not going to be rebooted.

--

James
-------
http://radio.weblogs.com/0112098/

Reply via email to