On 4/18/07, James Strachan <[EMAIL PROTECTED]> wrote:
On 4/17/07, Alvaro Carrasco <[EMAIL PROTECTED]> wrote:
>
> >
> > Its purely a case that the Stomp+JMS came later. We really could do
> > with revving Stomp to 1.0 or 1.1 to tidy up a few headers we'd used.
> >
> > I think the latest ActiveMQ should support the cleaner header names
> > from Stomp+JMS (if it doesn't we can fix it easily).
>
> Fair enough.
>
> Would it be a good idea to actually somehow embed StompConnect?
> That way we're not duplicating effort...
It might be - it'd certainly be less code to maintain; though the
native Stomp support in ActiveMQ is at a lower level; so can be more
efficient (e.g. not requiring a thread per session etc).
BTW one other difference is StompConnect only exposes pure JMS
features; whereas the native support of Stomp in ActiveMQ exposes
pretty much all of the features in ActiveMQ as extensions
--
James
-------
http://radio.weblogs.com/0112098/
