John, It indeed works as the old module did. Igor explained what it was used for and I made sure it still works the same. The main usecase is a component on a single page instance which requires a conversation. The conversation should live as long as the component is on the page.
We also discussed if ConversationalComponent should be changed into an annotation. I think it should have been an annotation from the start, but changing it now would force users to change their code only for aesthetics. The interface works fine, it only looks nicer if it were an annotation. So we decided to keep it as is. Best regards, Emond On Friday 10 January 2014 09:47:46 John Sarman wrote: > Emond, > Also it is very possible I misinterpreted how cdi-1.0 worked. I personally > believe that the auto conversation that is started should only end if the > propagation rule fails, that is navigate to a new page that is a > bookmarkable when propagation is set to NonBookmarkable, I read through the > cdi-1.0 code and it is basically cut and paste to the rewrite and placed at > the level so that the conversation is active without needing to use weld > internals to activate it :) :) After reading it I may have mistakenly > assumed that cdi-1.0 behaved like I described, I have not tested that. If > that is the case should we bring up a discussing to formally define what > semantic controls propagation. That is if you choose to have wicket-cdi > manage the conversation then every page needs to implement > ConversationalComponent otherwise the conversation will be ended even if it > passes the propagation criteria. If you choose to manually begin and end > conversations, then only the propagation criteria dictates conversation > propagation. > > Also we discussed removing the ConversationalComponent and making that an > Annotation. Is that still on the table? > > I am very impressed that you were able to remove the required dependency of > the cdi implementation. Glad you took the lead and made that happen. > > Thanks > John > > On Fri, Jan 10, 2014 at 8:59 AM, John Sarman <[email protected]> wrote: > > Should the Conversation be ended immediately? So if someone uses the > > ConversationalComponent to automate the conversation.begin(), does that > > override the propagation if the next page that is navigated to does not > > > > implement ConversationalComponent? I am just trying to understand the new > > > > way of managing the Conversation as it differs from the Wicket Cdi-1.0 > > pattern. In the 1.0 if a conversation was active and you navigated to a > > non bookmarkable page (assuming CdiConfiguration is configured with the > > default nonbookmarkable) then the Conversation would propagate > > regardless > > if the conversation was automatically managed or manually began. In 1.0 > > if > > the conversation was managed by the wicket-cdi and you navigated to a > > bookmarkable page then wicket-cdi would not propagate the Conversation and > > call conversation.end(). > > > > Thanks, > > John > > > > > > > > On Fri, Jan 10, 2014 at 8:49 AM, Emond Papegaaij < > > > > [email protected]> wrote: > >> The conversation is propagated, but also ended immediately. The > >> TestNonConversationalPage is rendered with count = 3, but this count is > >> lost on the next increment (rendering of the page is a different request > >> as > >> clicking the increment link). > >> > >> On Friday 10 January 2014 08:38:24 John Sarman wrote: > >> > Emond, > >> > I understand that. However whether the Conversation is began > >> > >> manually or > >> > >> > automatically should not dictate how the Conversation is propagated. > >> > In > >> > the TestConversationalPage starts the conversation via the > >> > ConversationalComponent interface then increments the counter. The > >> > >> counter > >> > >> > increment does not call setResponsePage(). When the next Link is > >> > >> clicked, > >> > >> > setResponsePage(new TestNonConversationalPage()); is fired. This is a > >> > > >> > non-bookmarkable page, but the Conversation is not propagated. This > >> > >> means > >> > >> > that the ConversationPropagation is not working. > >> > > >> > Thanks, > >> > John > >> > > >> > > >> > > >> > On Fri, Jan 10, 2014 at 3:02 AM, Emond Papegaaij > >> > >> <[email protected] > >> > >> > > wrote: > >> > > > >> > > Hi John,
