On Mon, Jan 19, 2009 at 10:37 PM, Ville Virtanen
<ville.virta...@cerion.fi>wrote:

>
> It sounds like an addition that would be good to have when implementing
> non-standard persistence strategies.


Thanks. Haven't heard from Howard but I'll open an issue anyway. I can get
this done without this enhancement, but it makes sense to me.

Ps. I'm very interested in the results Kalle if you care to share? (Or
> Kristian for that matter. Would it be possible to add a project to
> tapestry.formos.com?)


The current plan is to make it available as a submodule of Trails (
http://markmail.org/message/htv4tdxidex4xpv5#query:trails-dev%20conversation+page:1+mid:4pjnurcvwnw3untq+state:results).
But don't hold your breath - taking it from a prototype to a generic,
publicly available library including documentation will take some time. I
just quickly browsed through Kristian's implementation and the major
difference between his and mine is that his is more generic akin to Seam's
while mine is specifically conversation-within-page. But because of this
convention/requirement (in the latter), tracking conversations and ending
them automatically when they are abandoned or become idle is much easier.

Kalle



> Kalle Korhonen-2 wrote:
> >
> > Hey Kristian, thanks a lot for sending your project; it'll be interesting
> > to
> > compare the implementations. Yes, of course there's a million different
> > ways
> > of solving the problem but since the Tapestry architecture and services
> > offer a lot of functionality already for this, I tried to reuse most of
> > it.
> > And except for this small part, it worked out very well. Howard, care to
> > weigh in your opinion - would it not make sense to allow to selectively
> > discard changes of specific persistent strategies?
> >
> > Kalle
> >
> >
> > On Mon, Jan 19, 2009 at 3:14 AM, Kristian Marinkovic <
> > kristian.marinko...@porsche.co.at> wrote:
> >
> >> hi kalle,
> >>
> >> i've tried to solve this  with another indirection :).
> >>
> >> i encapsulate all conversation specific data in a ConversationContext
> >> object.
> >> if i want to invalidate a conversation i just remove this context.
> >>
> >> i'm going to send you a zip with my experimental project
> >> to your gmail address. maybe it helps.
> >>
> >> g,
> >> kris
> >>
> >>
> >>
> >>
> >> Kalle Korhonen <kalle.o.korho...@gmail.com>
> >> 19.01.2009 06:36
> >> Bitte antworten an
> >> "Tapestry users" <users@tapestry.apache.org>
> >>
> >>
> >> An
> >> Tapestry users <users@tapestry.apache.org>
> >> Kopie
> >>
> >> Thema
> >> Discard persistent field changes for one strategy only?
> >>
> >>
> >>
> >>
> >>
> >>
> >> I would like to discard all changes for one strategy only.
> >> PersistentFieldManager always iterates over all strategies when calling
> >> discardChanges() and I don't see any (easy) way to decorate the
> >> PersistentFieldManager to allow it to discard selectively for some
> >> strategies only. Would it be possible to add a discardChanges(String
> >> strategyName, String pageName) operation to PersistentFieldManager or to
> >> make the getStrategy(String strategyName) public? Or is there any other
> >> way
> >> to get a hold of a strategy in a service that the strategies haven't
> been
> >> contributed for if I know the type? This functionality would be needed
> >> for
> >> conversational scope implementation. At least in my mind it makes
> perfect
> >> sense to allow this, I'll open an issue if somebody agrees. For now, I
> >> can
> >> make it work by contributing the same
> ConversationPersistentFieldStrategy
> >> to
> >> another service in addition to PersistentFieldManager in my AppModule,
> >> but
> >> I
> >> can't make it "plugn'n'play" without the "selective discard"
> >> functionality.
> >>
> >> Kalle
> >>
> >>
> >
> >
>
> --
> View this message in context:
> http://www.nabble.com/Discard-persistent-field-changes-for-one-strategy-only--tp21537228p21557882.html
> Sent from the Tapestry - User mailing list archive at Nabble.com.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscr...@tapestry.apache.org
> For additional commands, e-mail: users-h...@tapestry.apache.org
>
>

Reply via email to