|
Re: OCL Checkers [message #54885 is a reply to message #54692] |
Sun, 04 May 2008 14:40 |
Eclipse User |
|
|
|
Originally posted by: give.a.damus.gmail.com
Hi, Mark,
See some comments in-line, below.
Cheers,
Christian
On Friday 05-02-2008 (09:58), Mark Melia wrote:
> Hi all,
> I have a general question about how OCL checking works, as far as I
> can see OCL checking in EMF is checked by iterating through the
> constraints and testing them against the instance model. Therefore if
> I have a number of constraints all the information that the OCL
> checker can give me is whether a constraint passed or failed.
That is the simplest approach to implement with the current MDT OCL
API, yes.
> Some of the failed constraints could have a consequential effect on
> the rest of the constraint - fix one constraint and some other
> constraints would also be fixed. Does any OCL checker check for this?
I don't know of any that are based on MDT OCL, but it's quite
conceivable that the failure of a constraint could be broken down to
find the particular sub-expression whose value caused the overall
constraint value to be false. Using this information, then, other
constraints could be analyzed to see whether they contain similar
sub-expressions that cause them to fail for the same reason.
> Is there any OCL checker which analysis the constraint violation,
> possibly classifying them - identifying for example consequential
> constraint violations?
I don't know of any such checker based on MDT OCL. I invite any who
know of one (or have implemented one) to reply!
> Sorry for the general question, but I do think this will be of
> interest to the list.
No apologies needed -- of course it is interesting. Everything
OCL-related is :-) In fact, one of the benefits of using OCL for the
specification of constraints is its simplicity and, therefore, the
relative ease with which such analyses can be performed (as compared
to other languages). It should be feasible, for example, to
automatically determine that some constraint is contradictory with
another, or that it implies another. This kind of higher-order
validation of constraints is an area that that the MDT OCL component
has not yet considered, but which would be a good fit for the tooling
mandate of the OCL Tools component.
> Mark
--
I'm trying a new usenet client for Mac, Nemo OS X.
You can download it at http://www.malcom-mac.com/nemo
|
|
|
Powered by
FUDForum. Page generated in 0.03476 seconds