On Sunday, December 4, 2011 8:31:48 PM UTC+1, vehdra music wrote:
>
> On Dec 3, 9:53 pm, Thomas Broyer <[email protected]> wrote:
> > If your 3 parts are tightly linked (you have different actions and
> > different filters for each different "main content"), given that they're
> > displayed next to each other, then you only need one activity.
> >
> > I'd however try to code them as three distinct "components" (widgets)
> > linked only through events (true events with event-handlers using
> > addHandler, or simply using callbacks).
>
> So, for example I will have to create with (or without) UIBinder a
> widget for every actions menu or filters menu that I need (dispite of
> reusing anyone when I could). For example UsersListActions widget and
> UsersListsFilter widget.
>

Hard (impossible?) to answer without knowing more about your specific 
needs, but do the Simplest Thing That Could Possibly Work™

On more doubt. Wich one would be the way to communicate between this
> widgets (actions and filters panels) and the current presenter (in my
> case, the current activity)?
>
Your widgets are part of the view, and your presenter is, well, the 
presenter for the view (you can segregate the widgets behind "sub 
interfaces").
 

>  That should make it easier maintain
> > the whole thing (otherwise your "main content with actions and filters"
> > could grow and become unmaintainable) but more importantly, if you think
> > you could visually separate them later, that would make it easier (in 
> that
> > event, route the events through the event bus and "you're done"). It 
> would
> > make it possible/easier to reuse one part in different views (e.g. if
> > most/all lists have the same set of actions).
> >
> > I'm afraid there's no "one size fits all" approach; it (in part) depends
> > how you imagine your app will evolve. The whole idea of activities are to
> > decouple things, it doesn't make sense for things that are tightly 
> coupled
> > (unless they're separated visually into non-adjacent areas). The idea is
> > that, for instance, a "main menu", a "list of things" and "details about
> > one thing" could all appear on the screen at the same time on a desktop,
> > but appear as "sequential screens" on a smart phone (and using MVP within
> > an activity, you can in addition decouple the view –wide on a desktop,
> > narrower on a smartphone– from the behavior)
> >
> > To me, however, your current design is clearly not the best (or i
> > misunderstood it): why use MainDisplayFilters and MainDisplayActions as
> > singletons that you clear/populate each time instead of simply using
> > distinct instances in each view?
>
> I did it in this way because I don't know how can I tell to my
> presenter from some widget to perform some action (ie: delete some
> users) or how to communicate to the vie.
>
> For example, on the start method of UserListActivity I make a rpc to
> load the users and populate my UsersListViewImpl. IE: How can I tell
> to my UsersListsActions widget that there are n users? Or how can I
> know from my UsersListsActions widget how many rows are selected in my
> UsersListViewImpl? How can I tell from my UsersListsActions to my
> UserListPresenter to delete some users?
>
What I understood is that you're using singletons and when you need them, 
you tell them "forget everything, and know this is what you should 
know/do". I'm just saying that instead of singletons and "forget 
everything", you should create instances that the activity/view *owns* and 
are initialized once and for all.

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/google-web-toolkit/-/pNLumTohzFwJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/google-web-toolkit?hl=en.

Reply via email to