[news.eclipse.tools.emf] Re: RecordingCommand causes headache

[Mario Winterer <mario.winterer@xxxxxxx>]

Christian W. Damus schrieb:
> Hi, Mario,
>
> This is good news!  Why do you say "Oh, no?"  Is there something else?

No, no, everything's fine - its just because my poor unit test has 
nothing to do now... ;-)

> :-)
>
> Mario Winterer wrote:
> > Oh no!
> > My unit test case works with EMF Transaction Core 1.2.1 (Eclipse 
> > 3.4.1). Yesterday, the same test failed with EMF Transaction Core 
> > 1.2.0 (Eclipse 3.4.0). Your bugfix seems to work...
> >
> > Christian W. Damus schrieb:
> > > Hi, Mario,
> > >
> > > Yes, that is what I meant.  Sorry to be so opaque.  :-)
> > >
> > > cW
> > >
> > > Mario Winterer wrote:
> > > > Christian W. Damus schrieb:
> > > > > I, Mario,
> > > > >
> > > > > Yes, this JUnit test looks like a good Bugzilla contribution.   :-)
> > > >
> > > > Should I file a bug for this purpose?
> > > >
> > > > > I have already fixed one bug in the undo/redo of RecordingCommands 
> > > > > that are used as triggers, but I think more work is needed.  
> > > > > Probably, the RecordingCommand will need to be able to create its 
> > > > > own nested transaction, as the AbstractEMFOperation can do, in the 
> > > > > appropriate circumstances.
> > > > >
> > > > > Cheers,
> > > > >
> > > > > Christian
> > >
> > > -----8<-----