Re: OPENEJB-1099 - Reentrant calls to Stateful beans (was: @AccessTimeout)

[David Blevins <da...@visi.com>]

On Jun 7, 2010, at 1:03 PM, Thiago Veronezi wrote:

> Thanks, David!
> I'm pretty excited: My first apache code has been committed! :O)
> I updated my code for the OPENEJB-1099 using your changes. A new Patch is
> available...
> 
> https://issues.apache.org/jira/secure/attachment/12446516/OPENEJB-1099.patch
> 

Excellent!  Got that committed.  

I've added you to the contributors group in JIRA and marked these as done by you:

  OPENEJB-1099: Reentrant calls to @Stateful beans unsupported
  OPENEJB-1144: @Stateful javax.ejb.AccessTimeout
  OPENEJB-1145: @Stateful javax.ejb.ConcurrentAccessTimeoutException
  OPENEJB-1209: <stateful-timeout> element
  OPENEJB-1210: <access-timeout> element (non-writable)
  OPENEJB-1211: @javax.ejb.StatefulTimeout overriding via <stateful-timeout>
  OPENEJB-1212: @javax.ejb.AccessTimeout overriding via <access-timeout>

Not sure if you also wanted:

  OPENEJB-1146: javax.ejb.StatefulTimeout


Went ahead and added an AccessTimeout property to the Stateful container so we can have a configurable default value if the bean didn't supply one.  Figured it'd be a nice descriptive looking diff we could use as a reference in case we needed to do something similar.

In terms of supporting AccessTimeout, we still need to work in all the spec rules.  Check out the EJB 3.1 spec.  Basically -1, 0, and 1+ all have different rules.

For the test case, it'd be great if we could get one that tests the reentrancy.

Pretty fantastic work so far.  We're all really cooking on these EJB 3.1 features.  It's pretty darn exciting!

If we keep going like this, we might be second in line to certify!


-David