Eclipse Community Forums: EMF » Bug in validation with ecore model in 3.5 : "containment or bidirectionnal reference must be un

Home » Modeling » EMF » Bug in validation with ecore model in 3.5 : "containment or bidirectionnal reference must be un

Show: Today's Messages :: Show Polls :: Message Navigator

Bug in validation with ecore model in 3.5 : "containment or bidirectionnal reference must be un [message #431430]

Thu, 09 July 2009 09:51

Steven Derrien

Messages: 50
Registered: July 2009

Member

Hello,

I recently switched to Galileo and I am now facing some strange bug
during the model validation stage :

I get a "containment or bidirectionnal reference must be unique" error
for a EReference field whereas my model passed validation without
trouble in 3.4.

I am not sure what the meaning of this validation error is, anyhow, this
error can be simply avoided by recreating in the editor the very same
Referance field (same name, same containment and upperbound properties).

This stringly suggests that is a bug rather than a more restrictive
validation in 3.5.

Is anybody aware of this issue ?

Thanks in advance,

Steven

Report message to a moderator

Re: Bug in validation with ecore model in 3.5 : "containment or bidirectionnal reference must b [message #431431 is a reply to message #431430]

Thu, 09 July 2009 10:08

Ed Merks

Messages: 30691
Registered: July 2009

Senior Member

Steven,

Comments below.

Steven Derrien wrote:
> Hello,
>
> I recently switched to Galileo and I am now facing some strange bug
> during the model validation stage :
>
> I get a "containment or bidirectionnal reference must be unique" error
> for a EReference field whereas my model passed validation without
> trouble in 3.4.
That's because of https://bugs.eclipse.org/bugs/show_bug.cgi?id=259368
>
> I am not sure what the meaning of this validation error is, anyhow,
> this error can be simply avoided by recreating in the editor the very
> same Referance field (same name, same containment and upperbound
> properties).
>
> This stringly suggests that is a bug rather than a more restrictive
> validation in 3.5.
Nope.
> Is anybody aware of this issue ?
Definitely. You just need to change your feature to set "unique" to
false. The generator "auto corrects" this anyway, which is why you
never noticed it's a problem.
>
> Thanks in advance,
>
> Steven

Report message to a moderator

Previous Topic:	[CDO] Need help with the testing framework
Next Topic:	Inheritance of EOperations

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

]

Current Time: Wed Dec 11 21:33:18 GMT 2019

.:: Contact :: Home ::.

Breadcrumbs

Sign up to our Newsletter