[ 
https://issues.apache.org/jira/browse/CAUSEWAY-3779?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17863866#comment-17863866
 ] 

Daniel Keir Haywood commented on CAUSEWAY-3779:
-----------------------------------------------

[~hobrom]  - just to mention on this ... with the recent additions I made to 
`PageRenderSubscriber` and `MetricsService`, it means that in 
`PageRenderSubscriber#onRenderedDomainObject` we can use the `MetricsService` 
to tell us which objects (as bookmarks) were loaded in order to render a given 
page.

My thinking is that we could cache this information somewhere, and then use it 
in `PageRenderSubscriber#onRenderingDomainObject` to pre-load a bunch of domain 
objects quickly.  We would likely need to partition them by logical type name, 
but even so it would save a lot of time.

For example, on the Estatio home page I observe >2000 objects being loaded.  If 
there are 20 different entity types, say, then that could become just 20 
database calls.

For this ticket, what we need is an API that can do that efficient bulk 
loading.  Then we just have to wire the two things together.

> Support for Query By Id(s)
> --------------------------
>
>                 Key: CAUSEWAY-3779
>                 URL: https://issues.apache.org/jira/browse/CAUSEWAY-3779
>             Project: Causeway
>          Issue Type: Improvement
>          Components: Persistence JDO, Persistence JPA
>            Reporter: Andi Huber
>            Assignee: Andi Huber
>            Priority: Major
>             Fix For: 3.2.0
>
>
> Given some stringified Entity IDs, fetch me all the entities with those IDs 
> in one bulk.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to