Hi,
I'd like to open a new thread to share my thoughts on event handling in slide.
I would like to start working on this issue at the beginning of the next year.
My basic thoughts are:
- Events should be fired for every operation that could be of interest for any 
recipient.
- There should be two types of events: Evens that are fired inside 
transactions (vetoable events) and events that are distributed as a bundle at 
the end of a transaction. So events should be able to replace interceptors 
(that are currently not transaction-aware).
- The events that are fired should contain a reference to the content that it 
belongs to. So it should be possible to check the content of a document 
inside a transaction if the vetoable write event occurs and deny it if 
needed. Btw: There should be a way to read the stream of a resource without 
confusing slide (so that slide can read from the stream again to write the 
resource).
- The events should be extendable so that custom events can be fired and 
distributed in the same way as "core" events.

The events should be able to connect parts of slide without coupling them too 
tightly. Think of some kind of presentation layer with caching, integration 
of lucene or a workflow engine.
In a first step events are fired inside the VM, maybe some event listeners 
could be used distribute them via JMS later to enable clustered environments.
I'd like to use naming conventions that are similar to swing events (events, 
listeners, adaters...) because everything should sound familiar and I think 
swing/awt is the place where events are used most frequently.

Daniel

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to