Re: A CoreData Limitation?
Re: A CoreData Limitation?
- Subject: Re: A CoreData Limitation?
- From: mmalcolm crawford <email@hidden>
- Date: Tue, 17 May 2005 14:55:06 -0700
On May 17, 2005, at 2:49 PM, Nicko van Someren wrote:
On 17 May 2005, at 22:13, mmalcolm crawford wrote:
On May 17, 2005, at 12:40 PM, Nicko van Someren wrote:
Then you make the Row object a custom subclass of NSManageObject
and implement a custom implementation of various bits of key-
value coding so that valueForKey: @"foo" actually does a fetch
from the set of cells with a predicate of cell.column.name ==
"foo" and insertion with setValue:forKey: does the right thing too.
If you want to follow this pattern (and I'm still not sufficiently
clear on the actual task to advise on that) then the appropriate
mechanism here would be to use a fetched property: <http://
developer.apple.com/documentation/Cocoa/Reference/CoreData_ObjC/
Classes/NSFetchedPropertyDesc.html>
Yes, I thought about that. The problem is that the orginal request
seemed to imply that columns needed to be dynamically created and
deleted whereas fetched properties are attributes of the model
entities and these can not be modified once there are any instances
of the entities. Thus it does not look to me like you can create a
fetched properties for a new column when it gets created.
If this is the case, then it's not clear that Core Data is an
appropriate solution -- by overriding valueForKey: in the way you
propose you're basically subverting the idea of having a model.
Bill's follow-up addresses the issue in more depth...
mmalc
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Cocoa-dev mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden