[SLING-11133] Cache model for its implementation type

• Type: Improvement
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Fix Version/s: Models Implementation 1.5.2
• Component/s: Sling Models
• Assignee: Dirk Rudolph
• Reporter: Dirk Rudolph
• Created: 10/Feb/22 10:40
• Updated: 03/May/22 12:13
• Resolved: 25/Feb/22 11:48

Description

Currently the ModelAdapterFactory caches a cacheable model for the requested type only

https://github.com/apache/sling-org-apache-sling-models-impl/blob/master/src/main/java/org/apache/sling/models/impl/ModelAdapterFactory.java#L430-L432

if (result.wasSuccessful() && modelAnnotation.cache() && adaptableCache != null) {
  adaptableCache.put(requestedType, new SoftReference<Object>(result.getValue()));
}

However, if a model is an adapter of multiple types an instance of it can also be returned for more than just the requested type.

interface AdapterType1 {}
interface AdapterType2 {}

@Model(adaptables = Resource.class, adapterType={AdapterType1.class, AdapterType2.class}, cache = true) 
class Model implements AdapterType1, AdapterType2 {}

assertSame(resource.adaptTo(Model.class), resouce.adaptTo(AdapterType1.class));
assertSame(resource.adaptTo(AdapterType1.class), resouce.adaptTo(AdapterType2.class));

While it is not save to cache the model for all of its adapter types (see SLING-11074), it is certainly save to cache the Model for its implementation type.

In fact, when caching the Model for the implementation type, caching it for the requested type is redundant. This is based on the assumption that the org.apache.sling.models.impl.AdapterImplementations#lookup() always returns the same implementation for given adapterType and adaptable.

Issue Links

• links to: sling-org-apache-sling-models-impl #35
• links to: sling-org-apache-sling-models-integration-tests #3