See below:
----- "kapil nayar" <kapilnayar1(a)gmail.com> wrote:
Hi Galder,
I looked at the preliminary writeup for the New Fine Grained
Replication API
Design at
http://community.jboss.org/wiki/newfinegrainedreplicationapidesign
When compared to the POJOCache does this design mandate an explicit
"commit"
(seems true because of the JPA style) to change the object instance
stored
in the cache/ replicated?
Yeah, commit is mandated as per JPA rules. It's at that point that we'll be able
to detect differences between objects in session and objects in cache.
If yes this design would NOT provide an
exact
replacement for the POJOCache style APIs and the included
flexibility.
Well, it's a totally different approach. This is not an attempt to replicate the
PojoCache APIs 100%, but instead provide programming model that users are more familiar
with and that is less error-prone than AOP based client APIs.
Could you clarify - did I miss something while reading the text.
Thanks,
Kapil
_______________________________________________
infinispan-dev mailing list
infinispan-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/infinispan-dev