[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [Newsgroup Home]
[news.eclipse.modeling.mdt.uml2] Re: Strategy for loading commercial UML XMI 2.1 (MaicDraw)

FYI, a fix for this bug has now been committed.

Kenn

"Kenn Hussey" <Kenn.Hussey@xxxxxxxxxxxxxxx> wrote in message 
news:h0mfgi$ij$1@xxxxxxxxxxxxxxxxxxxx
> Dawid,
>
> This is indeed a bug in EMF; see 
> https://bugs.eclipse.org/bugs/show_bug.cgi?id=279686. I'll do my best to 
> get this fix into Galileo.
>
> Kenn
>
> "Kenn Hussey" <Kenn.Hussey@xxxxxxxxxxxxxxx> wrote in message 
> news:h0lnpd$g3m$1@xxxxxxxxxxxxxxxxxxxx
>> Dawid,
>>
>> I made some progress debugging your model, but ran out of time. I'll try 
>> to spend some more time on it today. As I mentioned to Rafael, the 
>> document actually parses but gets hung up while resolving forward 
>> references at the end.
>>
>> Kenn
>>
>> "Dawid Loubser" <dawidl@xxxxxxxxxxx> wrote in message 
>> news:h0819a$nle$1@xxxxxxxxxxxxxxxxxxxx
>>> Hi Kenn,
>>>
>>> Our investigation also leads us ot believe there is a bug in EMF, there 
>>> is no way a model with 2 classes should cause scalability problems.
>>>
>>> It gives me hope that you have managed to load MagicDraw-produced XMI in 
>>> the past, any insight you can provide on why my sample model hosted at
>>>
>>> http://projects.solms.co.za/temp/testProject.xmi
>>>
>>> does not load will be appreciated more than you can imagine. Posting to 
>>> this newsgroup was somewhat of a last resort, I have tried so many
>>> combinations of using EMF/UML2 in my stand-alone program that my head
>>> is basically spining.
>>>
>>> Our goal is actually to load this model as a source model for an ATL 
>>> transformation, but after having given up on that route (using their UML 
>>> model loaders) we figured that we shoudl first achieve simple loading of
>>> the model in EMF/UML2 stand-alone.
>>>
>>> kind regards,
>>> Dawid Loubser
>>>
>>>
>>>
>>>
>>> Kenn Hussey het geskryf:
>>>> Eike,
>>>>
>>>> So far my investigation suggests a bug in EMF, so I'm not sure the 
>>>> problem is related to the size/scalability of the model itself...
>>>>
>>>> Kenn
>>>>
>>
>>
>
>