Hi Thomas,
I know we do not have a scope for Flash but it really should have a proper scope.
And yes I am well aware that ConversationScope is in CDI and it really ought to be
in Faces. Flash (scope) has a quite different way of using it.
What resources? Have you allocated folks that I do not know about? I would love
to be able to allocate someone to something
😉 ….. in all seriousness this is
an open-source project and I am proposing to modularize the scopes and yes that
to me means I would be signing up for that work, hence why I am proposing it.
On your polling to get usage numbers …. You’ll hit confirmation bias almost
Immediately because you will get a very limited set of folks responding. I mentioned
it as a way of measuring as right now we really “know” only because of our own
observations.
All this work is for setting Faces up for a future where a) we have data to make
informed decisions and b) a structure that makes it a whole lot easier to go through
a deprecation/removal process and c) makes it easier for someone to pick up said
piece when we have removed it and they want to take it forward.
On the note of ui:repeat I really would want that one to disappear from history
as yes it is fraught with problems and has never formally been specified as each
Faces implementation could have delivered its own implementation that is not
compatible as there is no way to determine that.
Thanks!
Kind regards,
Manfred Riem
we don't even have a real scope for Flash. Also ConversationScope is inside CDI, not JSF.
Also isn't Flash something like RedirectScope?
Really, please don't get me wrong here, but i don't see a big benefit here.
We should put our scarce resources into something like this:
https://github.com/eclipse-ee4j/faces-api/issues/1499
This is something that prevents users since JSF early days to correctly use validators inside ui:repeat.
If we need an analysis of which JSF scopes/features are used, we should just do a poll.
I also have like 10+ customers applications here and could do a simple analysis of used scopes.
I can however say: Flash, Conversation, Flow are almost unused.
Request, Session, View are really heavily used; some uses DeltaSpike WindowScope for a "session/tab" usecase.
Hi Rudy,
A couple of things come to mind.
-
Each scope would only be included when the customer actually wants to use it
-
If you want to deliver an alternative implementation you can do so easily
-
It would make sure the scope is only using public Faces / CDI APIs
-
We could see how much each scope is used which helps to figure out where we should put our scarce resources go forward
And on your question of compliance that is a TCK testing aspect of things, which does not limit allowing this modular approach.
Thanks!
Kind regards,
Manfred Riem
What is the benefit of having each scope in a separate module when they all need to be available to become a compliant implementation?
Hi Arjan,
One could imagine a separate module for each of the following:
FlashScoped
ViewScoped
ConverationScoped
ClientWindowScoped (annotation is not there yet)
RedirectScoped (annotation is not there yet)
Thanks!
Kind regards,
Manfred Riem
Hi,
Great idea! Any proposal for the modules we should have?
Hi,
Now that modules are becoming more and more mainstream is it time to consider create a modular Faces runtime?
Let me know your thoughts!
Thanks!
Kind regards,
Manfred Riem
_______________________________________________
faces-dev mailing list
faces-dev@xxxxxxxxxxx
To unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/faces-dev
_______________________________________________
faces-dev mailing list
faces-dev@xxxxxxxxxxx
To unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/faces-dev
_______________________________________________
faces-dev mailing list
faces-dev@xxxxxxxxxxx
To unsubscribe from this list, visit
https://www.eclipse.org/mailman/listinfo/faces-dev