[ADFEMG-85] RegionController : For internal framework use only Created: 15/Jan/13  Updated: 15/Sep/13  Resolved: 15/Sep/13

Status: Closed
Project: adfemg
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Major
Reporter: Jan Vervecken Assignee: Unassigned
Resolution: Fixed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified



Please consider the documentation section "21.4.5 What You May Need to Know About Extending RegionController for Page Fragments" [1]
saying "... create a new class that implements the RegionController interface ...".

Also consider the API documentation for interface oracle.adf.model.RegionController [2]
saying "*** For internal framework use only ***"

  • (q1) Why is the interface oracle.adf.model.RegionController documented as "For internal framework use only"?

many thanks
Jan Vervecken

Comment by chriscmuir [ 21/Jan/13 ]

Thanks for lodging the issue Jan.

Bug 16195699 raised.

Comment by Jan Vervecken [ 22/Jan/13 ]

Thank you for the update Chris.

  • about "Bug 16195699 raised."
    • On My Oracle Support I have been able to find bug 16195699, "IS REGIONCONTROLLER FOR INTERNAL FRAMEWORK USE ONLY?".

Jan Vervecken

Comment by Jan Vervecken [ 08/Feb/13 ]


Currently bug 16195699 has "Status 13 - Doc Bug (Response/Resolution)" and some unpublished comments.
So, it seems that some documentation should change, but which documentation and how, is currently unclear.

(Note that bug 16195699 currently refers to "external customer forum: http://java.net/jira/browse/ADFEMG-81", which should be http://java.net/jira/browse/ADFEMG-85 )

Jan Vervecken

Comment by chriscmuir [ 12/Feb/13 ]

The hidden response was:

@ Please remove RegionController discussion from the docs. It is an internal
@ inegration class between controller and model and not expected to be replaced
@ via a custom implementation

I've had a chat with Frank over do customers have a legit reason for requiring to do this? Is undocumenting it just the easy way out, or is there
really no need to override regioncontroller? Frank believes he's only seen two valid use cases for it, and overall most uses of it are for the wrong reason. As such he's comfortable for the documentation to be removed however with an ER in place to cover the two reasons.

For the record the two valid use cases Frank identified:

1. Users want to invoke JavaScript on a certain point during loading a page fragment

2. I for myself used in in my Oracle Magazine 2012 article ("Master and Commander") to refresh content of an injected managed bean and to
register my dynamic contextual event handler.

As a result a (non published) ER has been raised 16171278 entitled "PROVIDE LIFECYCLE HOOKS FOR PAGE FRAGMENT LOADING". Note this ER cannot be published as it includes numerous mentions of a specific customer. However an exert:

@ A common requirement of ADF customers is to receive a load notification for
@ page fragments in regions to execute client code or client side JavaScript
@ .
@ <censored> needs to automatically launch a popup dialog
@ when a user navigates to a page fragment (the page fragment may be the
@ default activity in a task flow but can also come as a second or third).
@ .
@ Other customers attempt to invoke JavaScript on fragment load to perform UI
@ component manipulation or call to 3rd party scripts
@ .
@ What is missing in ADF is
@ .
@ 1) ... a lifecycle notification that informs a listener in a bean that the
@ view is ready (which means that all the UI components are instantiated and
@ ready for rendering on the view)
@ .
@ 2) ... access to the region component ID the page fragment shows in. This
@ information would be required to compose the client ID of a component in the
@ page fragment. Developers will know about the naming containers within the
@ page fragment but don't know about other naming containers it is contained
@ in.
@ .
@ .
@ Suggestion: The lifecycle hook should work like a system event in JSF 2 and
@ should provide an event object that allows you to tell the region a fragment
@ is contained in plus the phase that is executed


Comment by Jan Vervecken [ 12/Feb/13 ]

Thank you for the update Chris.

  • about "... remove RegionController discussion from the docs ..."
    • Thanks for sharing this feedback (from bug 16195699), for considering use-cases and for pointing to a related enhancement request (16171278).

Jan Vervecken

Comment by chriscmuir [ 15/Jul/13 ]

Doc bug 16195699 has been completed in the 12.1.2 guide. The section "21.4.5 What You May Need to Know About Extending RegionController for Page Fragments" which would now sit logically at section 27.4.5 has been removed. By inference the Java API which states the RegionController interface for internal use only wins out.

There has been no work done on ER 16171278 to date.

As this issue was tracking bug 16195699, please consider closing this ADF EMG issue. If you'd like to track ER 16171278 via the ADF EMG issue tracker please raise a separate issue.


Comment by chriscmuir [ 15/Sep/13 ]

No further updates, closing issue.


Generated at Fri Sep 04 18:09:54 UTC 2015 using JIRA 6.2.3#6260-sha1:63ef1d6dac3f4f4d7db4c1effd405ba38ccdc558.