Re: Coredata and faulting
Re: Coredata and faulting
- Subject: Re: Coredata and faulting
- From: email@hidden
- Date: Tue, 29 Nov 2005 20:56:28 -0800
Scott Stevenson wrote:
On Nov 29, 2005, at 2:35 PM, email@hidden wrote:
I have a question concerning the implications of implementing search
in a coredata application containing a potentially large data set.
When searching a particular entity for say a string value, when
coredata
fetches the objects, does it un-fault the entire object or simply
the values
associated with the specific key?
My guess is that it would load data for all the simple attributes,
but it would not load in all of the relationships. Instead, those
keys themselves would be represented as faults.
Thats what I was hoping, though I wonder if relationships, even if
accessed would not un-fault more than just the relationship and not
non-requested attributes... I'm just wondering... perhaps thats a
cause of the performance that some are complaining about?
There are recommendations somewhere in the docs to consider caching
frequently-accessed remote values in local attributes to avoid some
of the overhead of joins.
Hm, I'll look into that.
Still, keep in mind that this behavior is fluid and could all
change in the future for optimizations (which is the whole point of
declarative design). Don't go overboard in designing your model
around one particular type of algorithm. Premature optimizations...
yadda yadda.
Right, just not wanting to run into a wall if I can see it coming...
Thanks for the response.
Andre
_______________________________________________
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