[weld-dev] Should you allow starting of an external container from mvn test?

[Steven Boscarine]

Agreed.  There's nothing wrong with giving people a lot of power, 
especially when you document the best practices behind them.

This just happens to be a feature I could never allow to be used in my 
organization as part of the default test group.  For our organization, 
the paradigm is that "mvn clean install" doesn't have external 
dependencies and will not execute code outside of it's own project.  For 
others shops, this may be a great way to write their tests.

For me, the upcoming weld standalone and embedded JBoss AS test 
implementations are much more compelling and I can't wait to try them out.


On 01/05/2010 03:10 PM, Dan Allen wrote:
> I still say that as long as the behavior is easily tuned (enabled, 
> disabled, proper instance location) and sufficiently documented, we 
> should offer the flexibility for the remote container to be started 
> automatically.
>
> -Dan
>
>