| beginner questions: what is best practice to implement EMF derived methods with OCL [message #37651] |
Mon, 24 September 2007 01:47  |
Eclipse User |
|
|
|
Hi, Everybody.
I selected available tutorials on OCL and the EMF validation framework,
and I plan to go through the material to learn how to use OCL
to implement EMF derived methods.
Has someone a specific hint to best practice how to implement EMF
derived methods with OCL?
Or has someone a hint in which order I shall do things?
Or a hint, which versions to use? I installed the latest maintenance
builds of everything, but are there reasons to go for the latest
milestones and integrations?
All I am looking for now is an easy start. We wont go live for months.
Best Regards, Philipp Kutter
|
|
|
| Re: beginner questions: what is best practice to implement EMF derived methods with OCL [message #37973 is a reply to message #37651] |
Mon, 24 September 2007 10:10  |
| Eclipse User |
|
|
|
Originally posted by: cdamus.ca.ibm.com
Hi, Philipp,
Using OCL to specify the bodies of EOperations isn't really in the scope of
either of EMF's validation frameworks (Diagnostician/EValidator or
ModelValidationService).
We always recommend the latest available stable builds (be they
maintenance-branch builds or stable HEAD-branch builds), of course, because
we created them just for you! ;-) I guess the most important factor is
whether you can live with updating dependencies in so many components: EMF
Core, EMF Query/Validation, OCL, even UML. Practically speaking, though,
there are no significant plans for the next OCL release (1.2) that will
improve the implementation of EOperations using OCL, so the 1.1.1 release
is a good place to start when it comes this week.
The article I linked to in my reply to your other post shows how you can
integrate OCL into your EMF code generation to provide EOperation bodies.
I can't really say, though, that it is a "best practice." It is just the
only practice that I know of so far, until the OCL Tools OCL-to-Java
transformation becomes available (sounds like it will in the time frame
that you're looking at).
The article is a working example, so hopefully it constitutes an easy start.
I think it's the easiest that I can offer ... perhaps others in the
newsgroup can share their materials, too.
HTH,
Christian
Philipp W. Kutter wrote:
> Hi, Everybody.
> I selected available tutorials on OCL and the EMF validation framework,
> and I plan to go through the material to learn how to use OCL
> to implement EMF derived methods.
>
> Has someone a specific hint to best practice how to implement EMF
> derived methods with OCL?
>
> Or has someone a hint in which order I shall do things?
>
> Or a hint, which versions to use? I installed the latest maintenance
> builds of everything, but are there reasons to go for the latest
> milestones and integrations?
>
> All I am looking for now is an easy start. We wont go live for months.
>
> Best Regards, Philipp Kutter
|
|
|
Powered by
FUDForum. Page generated in 0.02811 seconds
Home