Looking at two classes A and B, where B is dependent on one (constructor injection); B is registered as a single instrumentation in the container, and it has been made. A has been registered as an intent in a MultiTenantContainer, which is different on the tenant ID.
Question 1: Will
Resolve Identify that B should also be different on the tenant ID? Question 2: The same, but A has been registered as a lifetime - multi-tenant has been suppressed within the container and is dependent on B.
Func & Lt; Owned & lt; A & gt; & Gt; Update: Finish! This was a case of "what I wanted actually ..." because of the lack of understanding of my initial words; Specifically, I needed frequencies that were shared / shared in all the proposals of the tenant's lifetime , which was not in existence. Got some guidance, presented a patch on issue # 318; it was pulled in the trunk on the changes in 752, and we are good to go! Now the above questions will be read: Looking at two classes A and B, where B is dependent on one (constructor injection); B is registered in the container as instantperant , and it has been created. A is registered in the Multitant Container as Instant Advertisement , different on the Tenant ID ... And the answer to both questions is "Yes" now.
Thank you for helping and maintaining this great tool Travis and Nick.
If I correctly interpret my example that the answer to both numbers is "no".
Against any longer-lived example, Autofack is ever getting references for a short time example.
However this can mean that reconsidering ideas here and there, in my experience, makes it more predictable for behavior.
Hope it helps!
Nick
(BTW, if you need some input on a specific scenario that you might consider making another question with concrete components.) < / Div>
Comments
Post a Comment