On 12/27/06, Tres Seaver <[EMAIL PROTECTED]> wrote:
I'm not sure that is quite right -- the point of raising the exception
on move is to "back out" of any further work at that point, and up
through several layers of call stack. If we move to an event, then we
have to revisit all those layers, to ensure that the "termination path"
does no work after the event is emitted.
That's not 100% correct. The exception is explicitly called in
'_invokeWithNotification' and the exception arguments are acted on,
and passed to 'notifySuccess'.
My understanding here is that the point of raising the exception here
at the time was to overcome the lack of a established event framework.
I understand the point about code that expect to raise the exception,
but I don't see a reason that couldn't be fixed in a
backwards-compatible way, by both catching the exception and also
having a subscriber. Maybe that's too complicated though.
Sidnei da Silva
Enfold Systems http://enfoldsystems.com
Fax +1 832 201 8856 Office +1 713 942 2377 Ext 214
Zope-CMF maillist - Zope-CMF@lists.zope.org
See http://collector.zope.org/CMF for bug reports and feature requests