On May 30, 2008, at 6:39 PM, Mike Schrag wrote:
Now I'm just going to have to go look into fixing this in Wonder.
How much effort is expended being scared of inverse faulting (which
is a problem that seems like it really shouldn't be all THAT hard
to fix with a little elbow grease)?
It will definitely take a little more magic to pull off completely
and testing to see what the side-effects will be, but I have a
prototype that shows this is pretty straightforward to do. The one
catch is that includeObjectsInPropertyWithKey (or whatever that
method is) does a containsObject(..) call before adding. To answer
that question you have to fault the array. However, it appears to
work OK if I just make that assume false if it's not faulted at that
point. The downside of that is that willChange is called even
though once the array is eventually faulted, it may turn out that it
DIDN'T change (because it defers the uniqueness check until the
fault actually fires). So basically you get too many willChange
calls if you add duplicates to an unfaulted array. This doesn't
appear to actually be a problem in the limited testing I did,
though. Weekend time ...
willChange() gets called a lot so I don't see this as a serious
downside. I can't think of why it would cause problems as long as the
EOs are in an EC.
Chuck
--
Practical WebObjects - for developers who want to increase their
overall knowledge of WebObjects or who are trying to solve specific
problems.
http://www.global-village.net/products/practical_webobjects
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Webobjects-dev mailing list (Webobjects-dev@lists.apple.com)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/webobjects-dev/archive%40mail-archive.com
This email sent to [EMAIL PROTECTED]
