Hi,

On 3/29/06, Felix Meschberger <[EMAIL PROTECTED]> wrote:
> > OT: why does jcr-rmi have a dependency on commons-logging?
> As Jukka pointed out, it was added by me some time ago: This dependency
> was added because there is some logging in the event triggering
> implementation. If logging is removed, the dependency may be removed as
> well. Maybe, now that we use SL4J for logging, we might add a dependency
> to SL4J ?

SLF4J should be fine for now. In longer term I think we should avoid
the event pooling/polling mechanism in favor of remote callbacks (see
JCR-302) to make the RMI layer as transparent as possible. That would
also remove the need for logging in JCR-RMI. The callback approach
however has some issues related to the remote listener codebase, so
the issue is postponed.

BR,

Jukka Zitting

--
Yukatan - http://yukatan.fi/ - [EMAIL PROTECTED]
Software craftsmanship, JCR consulting, and Java development

Reply via email to