[news.eclipse.tools.emf] Re: [EMF Transaction] Notification list will no

["Christian W. Damus" <cdamus@xxxxxxxxxxxxx>]

Hi, Ben,

I would certainly be open to the IllegalStateException in the case of 
notifications without a transaction context while another transaction is 
in progress on another thread.  This is consistent with the case where 
there is no transaction at all.

By any chance, is the offending notification in your scenario one that 
is compatible with a read-only context, such as a proxy resolution?  The 
reason I ask is that, currently, there is no assertion that these must 
occur in a transaction context of any kind because we don't have a 
mechanism to assert transactions in the general case of reading the 
model content.

Please, do open a bugzilla.  We can explore further the options and 
their feasibility, there.  I expect that *something* will need to be done.

Thanks,

Christian

Ben wrote:
> Hi Christian,
>
> Your analysis is correct.  The TransactionUtils.writing method is as you 
> suggested.  We put in this workaround as an expedient solution, so you 
> may have noticed the race condition after (activeTransaction != null) 
> and before the notification gets recorded.
>
> Another reasonable approach would be to throw an IllegalStateException 
> (or perhaps a ConcurrentModificationException?) when a notification 
> comes in while a transaction is in play.  This would be similar to the 
> way that the IllegalStateException is thrown if the someone tries to 
> open a read-write transaction while in a read-transaction. What do you 
> think?  Also, please let me know if you want me to open a bugzilla for 
> this?
>
> Thanks!