Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Archived » M2M (model-to-model transformation) » [QVTO] Resolve on disjunctive mapping
[QVTO] Resolve on disjunctive mapping [message #854691] Tue, 24 April 2012 02:50 Go to next message
Christopher Gerking is currently offline Christopher Gerking
Messages: 36
Registered: April 2011
Member
I am trying to apply the resolveIn operation to a mapping that uses the disjuncts keyword. The result is always null. When I apply the resolve directly to the mappings that the original one delegates to, everything works fine.

Is there a specific reason that the trace does not support disjunctive mappings?

Report message to a moderator

Re: [QVTO] Resolve on disjunctive mapping [message #855096 is a reply to message #854691] Tue, 24 April 2012 10:14 Go to previous messageGo to next message
Ed Willink is currently offline Ed Willink
Messages: 3446
Registered: July 2009
Senior Member
Hi

Pending a more informed reply...

I find the various resolve operators slightly hard to understand and the
interactions of disjuncts and inherits sometimes surprising. No doubt
many of the problems are due to my poor understanding, but I suspect
there are some areas where an imperfect QVTo implementation (or
specification) aggravates misunderstanding. The vagueness of the
specification may require you to do some empirical discovery to
determine which of the available constructs work reliably and in accord
with your evolved understanding.

HTH

Regards

Ed Willink

On 24/04/2012 07:50, Missing name Mising name wrote:
> I am trying to apply the resolveIn operation to a mapping that uses
> the disjuncts keyword. The result is always null. When I apply the
> resolve directly to the mappings that the original one delegates to,
> everything works fine.
>
> Is there a specific reason that the trace does not support disjunctive
> mappings?

Report message to a moderator

Re: [QVTO] Resolve on disjunctive mapping [message #857300 is a reply to message #855096] Thu, 26 April 2012 08:36 Go to previous messageGo to next message
Christopher Gerking is currently offline Christopher Gerking
Messages: 36
Registered: April 2011
Member
I see that the specification is not very precise. But in this case, i think it would not hurt to change the implementation towards support for resolving on disjunctive mappings, since currently the result is always null. I think no one actually relies on this null result for his existing transformations. In addition, it does not sound too hard to implement, since the information about mapped objects only has to be propagated such that it is available for the disjunctive mapping. Bugzilla?

[Updated on: Thu, 26 April 2012 08:39]

Report message to a moderator

Re: [QVTO] Resolve on disjunctive mapping [message #857312 is a reply to message #857300] Thu, 26 April 2012 08:50 Go to previous message
Ed Willink is currently offline Ed Willink
Messages: 3446
Registered: July 2009
Senior Member
Hi

The Eclipse implementations endeavour to follow the specification, so if
you think the specification needs enhancement
http://www.omg.org/report_issue.htm is the appropriate channel.

Regards

Ed Willink

On 26/04/2012 13:36, Missing name Mising name wrote:
> I see that the specification is not very precise. But in this case, i
> think it would not hurt to change the implementation towards support
> for resolving on disjunctive mappings, since currently the result is
> always null. I think no one actually relies on this null result for
> his existing transformations. In addition, it does not sound too hard
> to implement, since the information about mapped objects only has to
> be propagated such that it is available for the disjunctive mapping.
> Bugzilla?

Report message to a moderator

Previous Topic:[qvto]how to run qvto stand-along
Next Topic:[QVTO] Cloning intermediate class objects
Goto Forum:
  

[ Syndicate this forum (XML) ] [ RSS ]

Current Time: Tue Oct 08 07:12:43 EDT 2013

Powered by FUDForum. Page generated in 0.02001 seconds