--
You received this message because you are subscribed to the ADF Enterprise Methodology Group (http://groups.google.com/group/adf-methodology). To unsubscribe send email to adf-methodolo...@googlegroups.com
All content to the ADF EMG lies under the Creative Commons Attribution 3.0 Unported License (http://creativecommons.org/licenses/by/3.0/). Any content sourced must be attributed back to the ADF EMG with a link to the Google Group (http://groups.google.com/group/adf-methodology).
Steven Davelaar| Technology
Director | Oracle Consulting Tel +31 30 669 8113 | Fax +31 30 669 9966 Oracle Nederland BV | Rijnzathe 6, 3454 PV De Meern, The Netherlands PO Box 147, 3454 ZJ De Meern,The Netherlands | KvK nr. 30096087 |
Oracle Expert Services | JHeadstart |
Regards,
Lynn Munsinger
ADF Product Management
In the past we mostly pointed developers to using stateless EJBs, but we
are looking to possibly default future behavior around stateful EJBs (a
little bit more aligned with the ADF way of working).
So what would be your preferences?
Thanks
Shay
Could you address the following to the development please:
1) Java bean is annotated with @PageFlowScope(name = "name of the page flow")
2) Mouse-left click on a bean class -> create data control -> data
control based on the bean is created. It's ok how it works in
JDveloper now, especially in release 2.
3) The factory takes care about @PageFlowScope annotation and returns
an instance of a bean from the scope specified.
The same principle might be applied to the standard JEE (JEE5 or JEE6)
annotations too.
The main point is: a factory implementation behind a data control is
aware of them and returns (creates or retrieves)
an appropriate instance.
To come back to the subject of the discussion: i would expect from the
factory behind a data control to take care about
@Stateless (mappedName="ejb/nameJNDI") annotation and to return the
appropriate instance.
Regards,
Donatas
2012/1/24 fnimphiu <frank.n...@googlemail.com>:
First, we do have two different developer guides based on two different sample applications, hopefully you all have seen the Java EE + ADF data control + ADF Faces one:
http://docs.oracle.com/cd/E24382_01/web.1112/e17272/toc.htm
Edwin, you suggested that there are different types of accessors created for different named queries - but I've confirmed that queries are treated the same for findAll or those with parameters, so you shouldn't be seeing this behavior - what version are you using? Or if I'm not understanding you, please send me the steps to replicate what you're seeing. Also, the ejb-link in the web.xml thing is probably a bug - I'll test that out and file a bug for that.
LOV support is slated for a future release. What would you think about alleviating the need for developers to define the LOV explicitly and just defaulting any foreign key to an LOV (for example when Employees and Departments entities are created, DeptNo would be created as an LOV with Departments.DeptNo values). The developer could then set the display name or tweak other settings as needed.
And we have implemented the ability for CDI to be used for validation rules and control hints (so Edwin, your getDomain() would be an accessor). This might be coming in a future release at a future time (vagueness intended). What functionality beyond control hints and validation rules (and LOVs) are you expecting the data control to pick up?
Additionally, we'll be (optionally) generating annotations for sequences in the "Create Entities from Tables" wizard. So the table and sequence_generator annotations (and the other related settings) will be configurable in the wizard. Are you all using this wizard? Or crafting your entities piecemeal?
Thanks for your thoughts on this, it's appreciated.
Lynn Munsinger
Oracle JDeveloper & ADF Product Management
The stateful case, whether using a stateful Session bean or a Java service facade, is the most straightforward for the user. The Create operation under an accessor, whether it be a top-level accessor like 'departmentsFindAll' or a detail accessor like the employeesList owned by a department, performs two tasks. It creates a new instance of the entity, using the entity's default constructor, and it adds it to the context master. In the case of a top-level accessor, this context master is the data control itself, and so it calls the persistEntity() method. For a true master-detail accessor, it calls the AddMethod associated with the accessor method on the JPA entity, such as Departments.addEmployees(). In a stateful case involving an EXTENDED persistence context (available for stateful session beans only), users can continue to edit the newly created instance and all modifications will be persisted when the user invokes the Commit operation.
What this all would mean for the developer is much more symmetry between the ADFbc world and the JPA-based data control.
Regards,