On Thu, 2002-11-21 at 11:14, Craig Miskell wrote:
> 3) I'm almost certain there will be some borderline cases that it
> doesn't handle, but for now, it's more robust than what was there before
> (IMHO). That sounds quite uppity, doesn't it? I assure you that's not
> the intention.
Indeed, I just found one (I don't think it worked before either) - a
reflexive relationship, e.g. Parent-child relationship such as a Group
which has subgroups (a group has toParent and toChildren relationships)
may not work, because the insert order sees that the two objects are in
the same entity and just arbitrarily orders them (not necessarily parent
first, unfortunately)
I'll think about how to handle this, but at the moment my brain is fried
from previous insert order efforts. Any suggestions (or offers to do
it) greatly appreciated. Fully specified algorithms also accepted :-)
Craig
This archive was generated by hypermail 2.0.0 : Wed Nov 20 2002 - 17:31:39 EST