An instance of
NSManaged represents a single “object space” or scratch pad in an application. Its primary responsibility is to manage a collection of managed objects. These objects form a group of related model objects that represent an internally consistent view of one or more persistent stores. A single managed object instance exists in one and only one context, but multiple copies of an object can exist in different contexts. Thus object uniquing is scoped to a particular context.
- iOS 3.0+
- macOS 10.4+
- tvOS 9.0+
- watchOS 2.0+
The context is a powerful object with a central role in the life-cycle of managed objects, with responsibilities from life-cycle management (including faulting) to validation, inverse relationship handling, and undo/redo. Through a context you can retrieve or “fetch” objects from a persistent store, make changes to those objects, and then either discard the changes or—again through the context—commit them back to the persistent store. The context is responsible for watching for changes in its objects and maintains an undo manager so you can have finer-grained control over undo and redo. You can insert new objects and delete ones you have fetched, and commit these modifications to the persistent store.
All objects fetched from an external store are registered in a context together with a global identifier (an instance of
NSManaged) that’s used to uniquely identify each object to the external store.
Managed object contexts have a parent store from which they retrieve data representing managed objects and through which they commit changes to managed objects.
Prior to OS X v10.7 and iOS v5.0, the parent store is always a persistent store coordinator. In macOS 10.7 and later and iOS v5.0 and later, the parent store may be another managed object context. Ultimately the root of a context’s ancestry must be a persistent store coordinator. The coordinator provides the managed object model and dispatches requests to the various persistent stores containing the data.
If a context’s parent store is another managed object context, fetch and save operations are mediated by the parent context instead of a coordinator. This pattern has a number of usage scenarios, including:
Performing background operations on a second thread or queue.
Managing discardable edits, such as in an inspector window or view.
As the first scenario implies, a parent context can service requests from children on different threads. You cannot, therefore, use parent contexts created with the thread confinement type (see Concurrency).
When you save changes in a context, the changes are only committed “one store up.” If you save a child context, changes are pushed to its parent. Changes are not saved to the persistent store until the root context is saved. (A root managed object context is one whose parent context is
nil.) In addition, a parent does not pull changes from children before it saves. You must save a child context if you want ultimately to commit the changes.
A context posts notifications at various points—see
NSManaged for example. Typically, you should register to receive these notifications only from known contexts:
Several system frameworks use Core Data internally. If you register to receive these notifications from all contexts (by passing
nil as the object parameter to a method such as
add), then you may receive unexpected notifications that are difficult to handle.
Core Data uses thread (or serialized queue) confinement to protect managed objects and managed object contexts (see Core Data Programming Guide). A consequence of this is that a context assumes the default owner is the thread or queue that allocated it—this is determined by the thread that calls its
init method. You should not, therefore, initialize a context on one thread then pass it to a different thread. Instead, you should pass a reference to a persistent store coordinator and have the receiving thread/queue create a new context derived from that. If you use
NSOperation, you must create the context in
main (for a serial queue) or
start (for a concurrent queue).
In macOS 10.7 and later and iOS v5.0 and later, when you create a context you can specify the concurrency pattern with which you will use it using
init. When you create a managed object context using initWithConcurrencyType:, you have two options for its thread (queue) association
For backwards compatibility, this is the default. You promise that context will not be used by any thread other than the one on which you created it. In general, to make the behavior explicit you’re encouraged to use one of the other types instead.
You can only use this concurrency type if the managed object context’s parent store is a persistent store coordinator.
Private queue (
Queue Concurrency Type
The context creates and manages a private queue.
Main queue (
Queue Concurrency Type
The context is associated with the main queue, and as such is tied into the application’s event loop, but it is otherwise similar to a private queue-based context. You use this queue type for contexts linked to controllers and UI objects that are required to be used only on the main thread.
If you use contexts using the confinement pattern, you send the contexts messages directly; it’s up to you to ensure that you send the messages from the right queue.
You use contexts using the queue-based concurrency types in conjunction with
perform. You group “standard” messages to send to the context within a block to pass to one of these methods. There are two exceptions:
Setter methods on queue-based managed object contexts are thread-safe. You can invoke these methods directly on any thread.
If your code is executing on the main thread, you can invoke methods on the main queue style contexts directly instead of using the block based API.
perform ensure the block operations are executed on the queue specified for the context. The
perform method returns immediately and the context executes the block methods on its own thread. With the
perform method, the context still executes the block methods on its own thread, but the method doesn’t return until the block is executed.
It’s important to appreciate that blocks are executed as a distinct body of work. As soon as your block ends, anyone else can enqueue another block, undo changes, reset the context, and so on. Thus blocks may be quite large, and typically end by invoking
You can also perform other operations, such as:
You are strongly discouraged from subclassing
NSManaged. The change tracking and undo management mechanisms are highly optimized and hence intricate and delicate. Interposing your own additional logic that might impact
process can have unforeseen consequences. In situations such as store migration, Core Data will create instances of
NSManaged for its own use. Under these circumstances, you cannot rely on any features of your custom subclass. Any
NSManaged subclass must always be fully compatible with
NSManaged (that is, it cannot rely on features of a subclass of