Nigel,
Seems like these javadocs are out of sync. I'll raise a separate issue.
http://download.oracle.com/otndocs/jcp/jms-2_0-fr-eval-spec/index.html
On Tue, Jun 25, 2013 at 5:07 AM, Nigel Deakin <nigel.deakin_at_oracle.com>wrote:
> Clebert,
>
>
> On 24/06/2013 20:36, Clebert Suconic wrote:
>
> say you created a JMSContext, and used setClientID....
> JMSContext ctx = ...
> ctx.setClientID("someID");
>
> Say, if later we call
> JMSContext ctx2 = ctx.createJMSContext();
>
>
> Note that this method has one parameter, a sessionMode.
>
> http://docs.oracle.com/javaee/7/api/javax/jms/JMSContext.html#createContext%28int%29
>
>
>
> What would be the expected value of ctx2.getClientID() ?
>
>
> The javadocs for JMSContext#createContext(int sessionMode) say that this
> "creates a new JMSContext with the specified session mode using the same
> connection as this JMSContext and creating a new session. "
>
> Since the two JMSContext objects will use the same underlying connection,
> they have the same clientID, same exception listener and same connection
> metadata.
>
> I'm assuming there will be some connection reused by the JMSContext and
> ctx.createJMSContext() would reuse the internal connection as we talked
> earlier. I'm just double checking if this is everybody's expectations...?
>
>
> That's certainly my intention. The reason for this method is to allow you
> to create a second session on the same connection. The main reason for
> doing this is to allow you to send or receive messages using more than one
> thread at a time (but still using the same connection).
>
> The above applies only to Java SE and the Java EE application client.
> Applications running in the Java EE web or EJB container are not allowed to
> create more than one session per connection.This restriction dates from
> Java EE 6 and probably earlier: see Java EE 6 section EE 6.7.
>
> For this reason, JMSContext#createContext(int sessionMode) is required to
> throw a JMSRuntimeException if called in the Java EE web or EJB containers.
> This is also why there isn't a no-parameter version of this method.
>
> Nigel
>
>
> Clebert
>
>
>