On Thu, Jun 5, 2008 at 1:49 AM, Bruce Snyder <[EMAIL PROTECTED]> wrote:
> On Thu, Jun 5, 2008 at 1:45 AM, Lars Heinemann
> <[EMAIL PROTECTED]> wrote:
>> On Thursday 05 June 2008 09:39:58 Bruce Snyder wrote:
>>> On Thu, Jun 5, 2008 at 1:35 AM, Lars Heinemann
>>>
>>> <[EMAIL PROTECTED]> wrote:
>>> > Bruce,
>>> >
>>> > there is one big backdraw on your suggestion.
>>> > The quality of the testing is very dependent to the quality of the mock
>>> > classes used. After running such tests you can never say that a component
>>> > works with either smx3 or smx4 as they were never really tested with the
>>> > containers itself.
>>>
>>> Yes, I do admit that true integration testing is not achieved. But
>>> honestly, the lack of testing and lack of consistency in the testing
>>> of components today needs to be rectified.
>>
>> --> I agree. But then it would be faster and imho more effective to spend
>> additional time in test improvements instead of dropping our current testing
>> and do complete new mockup tests.
>
> I'm not saying that we should drop the current testing. I guess we can
> address this via Maven profiles. We could split out the tests into
> separate directories based on their type, e.g., smx3 integration
> level, smx4 integration level and unit level.

Whoops, hit send to quickly. Then we just filter each test directory
in a separate profile.

Bruce
-- 
perl -e 'print unpack("u30","D0G)[EMAIL 
PROTECTED]&5R\"F)R=6-E+G-N>61E<D\!G;6%I;\"YC;VT*"
);'

Apache ActiveMQ - http://activemq.org/
Apache Camel - http://activemq.org/camel/
Apache ServiceMix - http://servicemix.org/
Apache Geronimo - http://geronimo.apache.org/

Blog: http://bruceblog.org/

Reply via email to