Question

I’m looking for a memory efficient way to take only one row from a fetch result set. This must be random.

I thought using [context countForFetchRequest:fetch error:nil]; and get an int random value between 0 and that and offset + limit the fetch to 1 result. But I can’t find whether or not it doesn’t allocate memory for each item it counts.

Is “count” a lightweight operation? Or does it need to instantiate objects in the context before being able to count them?

Answer 1

The documentation is somewhat unclear, but it includes the phrase “number of objects a given fetch request would have returned.” Furthermore, Core Data tends to make things like count very lightweight – entity instances, for example, allow you to call count to find out the number of objects on the end of a to-many relationship without instantiating all those objects or firing that fault. I’d say go for it, but profile it yourself – don’t optimize prematurely!